Language：English   Publishing type：Research paper (international conference proceedings)  

Conventional cell tracking methods detect multiple cells in each frame (detection) and then associate the detection results in successive time-frames (association). Most cell tracking methods perform the association task independently from the detection task. However, there is no guarantee of preserving coherence between these tasks, and lack of coherence may adversely affect tracking performance. In this paper, we propose the Motion and Position Map (MPM) that jointly represents both detection and association for not only migration but also cell division. It guarantees coherence such that if a cell is detected, the corresponding motion flow can always be obtained. It is a simple but powerful method for multi-object tracking in dense environments. We compared the proposed method with current tracking methods under various conditions in real biological images and found that it outperformed the state-of-the-art (+5.2\% improvement compared to the second-best).

DOI： https://doi.org/10.1109/CVPR42600.2020.00388

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/CVPR.2019.01288

Language：English   Publishing type：Research paper (scientific journal)  

Cell detection in microscopy images is essential for automated cell behavior analysis including cell shape analysis and cell tracking. Robust cell detection in high-density and low-contrast images is still challenging since cells often touch and partially overlap, forming a cell cluster with blurry intercellular boundaries. In such cases, current methods tend to detect multiple cells as a cluster. If the control parameters are adjusted to separate the touching cells, other problems often occur: a single cell may be segmented into several regions, and cells in low-intensity regions may not be detected. To solve these problems, we first detect redundant candidate regions, which include many false positives but in turn very few false negatives, by allowing candidate regions to overlap with each other. Next, the score for how likely the candidate region contains the main part of a single cell is computed for each cell candidate using supervised learning. Then we select an optimal set of cell regions from the redundant regions under nonoverlapping constraints, where each selected region looks like a single cell and the selected regions do not overlap. We formulate this problem of optimal region selection as a binary linear programming problem under nonoverlapping constraints. We demonstrated the effectiveness of our method for several types of cells in microscopy images. Our method performed better than five representative methods, achieving an F-measure of over 0.9 for all data sets. Experimental application of the proposed method to 3-D images demonstrated that also works well for 3-D cell detection.

DOI： 10.1109/TMI.2015.2391095

Authorship：Last author,　Corresponding author   Language：English  

Language：English  

Publisher：南江堂  

DOI： 10.15106/j_seikei76_640

CiNii Research

Language：English  

DOI： 10.1109/ISBI60581.2025.10980754

Authorship：Last author,　Corresponding author  

DOI： 10.1109/WACV61041.2025.00421

Authorship：Last author,　Corresponding author   Language：English  

DOI： 10.1109/EMBC53108.2024.10782204

Authorship：Last author,　Corresponding author   Language：English  

Language：English   Publisher：Medical Image Analysis  

Automatic image-based severity estimation is an important task in computer-aided diagnosis. Severity estimation by deep learning requires a large amount of training data to achieve a high performance. In general, severity estimation uses training data annotated with discrete (i.e., quantized) severity labels. Annotating discrete labels is often difficult in images with ambiguous severity, and the annotation cost is high. In contrast, relative annotation, in which the severity between a pair of images is compared, can avoid quantizing severity and thus makes it easier. We can estimate relative disease severity using a learning-to-rank framework with relative annotations, but relative annotation has the problem of the enormous number of pairs that can be annotated. Therefore, the selection of appropriate pairs is essential for relative annotation. In this paper, we propose a deep Bayesian active learning-to-rank that automatically selects appropriate pairs for relative annotation. Our method preferentially annotates unlabeled pairs with high learning efficiency from the model uncertainty of the samples. We prove the theoretical basis for adapting Bayesian neural networks to pairwise learning-to-rank and demonstrate the efficiency of our method through experiments on endoscopic images of ulcerative colitis on both private and public datasets. We also show that our method achieves a high performance under conditions of significant class imbalance because it automatically selects samples from the minority classes.

DOI： 10.1016/j.media.2024.103262

Web of Science

Scopus

PubMed

Language：English  

DOI： https://doi.org/10.1007/978-3-031-72120-5_35

Authorship：Last author,　Corresponding author   Language：English  

DOI： https://doi.org/10.48550/arXiv.2408.14130

Authorship：Last author,　Corresponding author   Language：English  

Language：English  

DOI： https://doi.org/10.1016/j.media.2024.103262

Publisher：Image and Vision Computing  

Medical image segmentation is an important task in medical imaging and diagnosis. Data augmentation can substantially improve the accuracy of medical image segmentation when the dataset has a small amount of medical images. However, the data augmentation methods for medical image are usually based on big models that require extensive search space. Furthermore, excessively complex models often have a heavy burden for the general healthcare organization or researcher. To address this problem, we propose a method of data augmentation that is simple to implement even for the general researcher and simple to transplant across various models. Here we introduce our new methods called KeepMask and KeepMix, which can be simply ported to a variety of models and provide high performance. These methods allow data augmentation without any effect on the target organ or lesion and can also be adapted to multi-class segmentation. KeepMask and KeepMix can not only perturb the background of an existing medical image but also add target organs that are not present to it and generate new images based on the image. In this paper, we performed our methods on both binary class datasets and multi-class datasets and obtained better performance. We conducted numerous experiments showing the predicted segmentation images using our proposed methods obtained more accurate boundaries.

DOI： 10.1016/j.imavis.2024.105056

Web of Science

Scopus

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Knee  

Background: This study proposed an automatic surgical planning system for high tibial osteotomy (HTO) using deep learning-based artificial intelligence and validated its accuracy. The system simulates osteotomy and measures lower-limb alignment parameters in pre- and post-osteotomy simulations. Methods: A total of 107 whole-leg standing radiographs were obtained from 107 patients who underwent HTO. First, the system detected anatomical landmarks on radiographs. Then, it simulated osteotomy and automatically measured five parameters in pre- and post-osteotomy simulation (hip knee angle [HKA], weight-bearing line ratio [WBL ratio], mechanical lateral distal femoral angle [mLDFA], mechanical medial proximal tibial angle [mMPTA], and mechanical lateral distal tibial angle [mLDTA]). The accuracy of the measured parameters was validated by comparing them with the ground truth (GT) values given by two orthopaedic surgeons. Results: All absolute errors of the system were within 1.5° or 1.5%. All inter-rater correlation confidence (ICC) values between the system and GT showed good reliability (>0.80). Excellent reliability was observed in the HKA (0.99) and WBL ratios (>0.99) for the pre-osteotomy simulation. The intra-rater difference of the system exhibited excellent reliability with an ICC value of 1.00 for all lower-limb alignment parameters in pre- and post-osteotomy simulations. In addition, the measurement time per radiograph (0.24 s) was considerably shorter than that of an orthopaedic surgeon (118 s). Conclusion: The proposed system is practically applicable because it can measure lower-limb alignment parameters accurately and quickly in pre- and post-osteotomy simulations. The system has potential applications in surgical planning systems.

DOI： 10.1016/j.knee.2024.03.008

Web of Science

Scopus

PubMed

Language：English   Publisher：Modern Pathology  

Several studies have developed various artificial intelligence (AI) models for immunohistochemical analysis of programmed death ligand 1 (PD-L1) in patients with non–small cell lung carcinoma; however, none have focused on specific ways by which AI-assisted systems could help pathologists determine the tumor proportion score (TPS). In this study, we developed an AI model to calculate the TPS of the PD-L1 22C3 assay and evaluated whether and how this AI-assisted system could help pathologists determine the TPS and analyze how AI-assisted systems could affect pathologists’ assessment accuracy. We assessed the 4 methods of the AI-assisted system: (1 and 2) pathologists first assessed and then referred to automated AI scoring results (1, positive tumor cell percentage; 2, positive tumor cell percentage and visualized overlay image) for final confirmation, and (3 and 4) pathologists referred to the automated AI scoring results (3, positive tumor cell percentage; 4, positive tumor cell percentage and visualized overlay image) while determining TPS. Mixed-model analysis was used to calculate the odds ratios (ORs) with 95% CI for AI-assisted TPS methods 1 to 4 compared with pathologists’ scoring. For all 584 samples of the tissue microarray, the OR for AI-assisted TPS methods 1 to 4 was 0.94 to 1.07 and not statistically significant. Of them, we found 332 discordant cases, on which the pathologists’ judgments were inconsistent; the ORs for AI-assisted TPS methods 1, 2, 3, and 4 were 1.28 (1.06-1.54; P = .012), 1.29 (1.06-1.55; P = .010), 1.28 (1.06-1.54; P = .012), and 1.29 (1.06-1.55; P = .010), respectively, which were statistically significant. For discordant cases, the OR for each AI-assisted TPS method compared with the others was 0.99 to 1.01 and not statistically significant. This study emphasized the usefulness of the AI-assisted system for cases in which pathologists had difficulty determining the PD-L1 TPS.

DOI： 10.1016/j.modpat.2024.100485

Web of Science

Scopus

PubMed

Language：English   Publisher：Nature Communications  

Cell states are regulated by the response of signaling pathways to receptor ligand-binding and intercellular interactions. High-resolution imaging has been attempted to explore the dynamics of these processes and, recently, multiplexed imaging has profiled cell states by achieving a comprehensive acquisition of spatial protein information from cells. However, the specificity of antibodies is still compromised when visualizing activated signals. Here, we develop Precise Emission Canceling Antibodies (PECAbs) that have cleavable fluorescent labeling. PECAbs enable high-specificity sequential imaging using hundreds of antibodies, allowing for reconstruction of the spatiotemporal dynamics of signaling pathways. Additionally, combining this approach with seq-smFISH can effectively classify cells and identify their signal activation states in human tissue. Overall, the PECAb system can serve as a comprehensive platform for analyzing complex cell processes.

DOI： 10.1038/s41467-024-47989-9

Web of Science

Scopus

PubMed

Language：English   Publishing type：Research paper (international conference proceedings)  

Language：English   Publisher：John Wiley & Sons Australia, Ltd  

潰瘍性大腸炎(UC)に対する既存の内視鏡スコアでは、内視鏡的所見の有無に基づき疾患の重症度を客観的に分類する。したがって、各カテゴリー内の疾患における重症度の差異を反映しない可能性がある。炎症性腸疾患(IBD)を専門とする内視鏡医は重症度を判定し、炎症の程度の全体的な印象を診断する。本研究の目的は、IBDを専門とする内視鏡医によるUCの内視鏡的重症度評価を正確に反映する人工知能(AI)システムを構築することである。ランキング畳み込みニューラルネットワーク(ranking-CNN)を、IBDを専門とする内視鏡医が作成した13826ペアの内視鏡画像のUC重症度に関する比較情報を用いて訓練した。訓練されたranking-CNNを用いて、UC内視鏡階調スケール(UCEGS)により重症度を判定した。AIによって診断されたUCEGSを用いた重症度評価と、Mayo内視鏡サブスコアとの間に不一致がないことを確認する為に、相関係数を計算した。4名のIBDを専門とする内視鏡医と、AIによるUCEGSを用いた画像評価の平均の相関係数も求めた。AIにより診断されたUCEGSとMayo内視鏡サブスコアとのSpearman相関係数は約0.89であった。IBD専門の内視鏡医とAIによる評価の結果の相関係数はすべて0.95超であった(P<0.01)。以上より、本研究で構築されたAIはUCの内視鏡的重症度をIBD専門の内視鏡医と同等に診断できることが示された。

Language：English   Publishing type：Research paper (scientific journal)  

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings  

The paper proposes a novel problem in multi-class Multiple-Instance Learning (MIL) called Learning from the Majority Label (LML). In LML, the majority class of instances in a bag is assigned as the bag's label. LML aims to classify instances using bag-level majority classes. This problem is valuable in various applications. Existing MIL methods are unsuitable for LML due to aggregating confidences, which may lead to inconsistency between the bag-level label and the label obtained by counting the number of instances for each class. This may lead to incorrect instance-level classification. We propose a counting network trained to produce the bag-level majority labels estimated by counting the number of instances for each class. This led to the consistency of the majority class between the network outputs and one obtained by counting the number of instances. Experimental results show that our counting network outperforms conventional MIL methods on four datasets.

DOI： 10.1109/ICASSP48485.2024.10448425

Web of Science

Scopus

Language：English   Publishing type：Research paper (scientific journal)   Publisher：npj Precision Oncology  

Prognosis after neoadjuvant chemotherapy (NAC) for osteosarcoma is generally predicted using manual necrosis-rate assessments; however, necrosis rates obtained in these assessments are not reproducible and do not adequately reflect individual cell responses. We aimed to investigate whether viable tumor cell density assessed using a deep-learning model (DLM) reflects the prognosis of osteosarcoma. Seventy-one patients were included in this study. Initially, the DLM was trained to detect viable tumor cells, following which it calculated their density. Patients were stratified into high and low-viable tumor cell density groups based on DLM measurements, and survival analysis was performed to evaluate disease-specific survival and metastasis-free survival (DSS and MFS). The high viable tumor cell density group exhibited worse DSS (p = 0.023) and MFS (p = 0.033). DLM-evaluated viable density showed correct stratification of prognosis groups. Therefore, this evaluation method may enable precise stratification of the prognosis in osteosarcoma patients treated with NAC.

DOI： 10.1038/s41698-024-00515-y

Web of Science

Scopus

PubMed

Publisher：Proceedings - International Symposium on Biomedical Imaging  

This paper brings attention to a new issue in the development of datasets for AI-based pathological image analysis: the domain shift problem stemming from the storage period until digital scanning. Pathological slides that were sliced and stained in the past may also be scanned to prepare a dataset for AI in addition to the newly stained tissue. However, it leads to changes in their appearance and causes domain shift. We propose a domain generalization method that leverages the storage period as sub-domains. Furthermore, we introduce the ordinal adversarial loss that can perform well for ordinal domain classes. The experiments show the effectiveness of using the storage period as a sub-domain and the ordinal adversarial loss.

DOI： 10.1109/ISBI56570.2024.10635214

Web of Science

Scopus

Publisher：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

The PD-L1 rate, the number of PD-L1 positive tumor cells over the total number of all tumor cells, is an important metric for immunotherapy. This metric is recorded as diagnostic information with pathological images. In this paper, we propose a proportion estimation method with a small amount of cell-level annotation and proportion annotation, which can be easily collected. Since the PD-L1 rate is calculated from only ‘tumor cells’ and not using ‘non-tumor cells’, we first detect tumor cells with a detection model. Then, we estimate the PD-L1 proportion by introducing a masking technique to ‘learning from label proportion’. In addition, we propose a weighted focal proportion loss to address data imbalance problems. Experiments using clinical data demonstrate the effectiveness of our method. Our method achieved the best performance in comparisons.

DOI： 10.1007/978-3-031-58171-7_12

Web of Science

Scopus

Publisher：Lecture Notes in Computer Science  

In this paper, we address the segmentation of tumor subtypes in whole slide images (WSI) by utilizing incomplete label proportions. Specifically, we utilize ‘partial’ label proportions, which give the proportions among tumor subtypes but do not give the proportion between tumor and non-tumor. Partial label proportions are recorded as the standard diagnostic information by pathologists, and we, therefore, want to use them to realize the segmentation model that can classify each WSI patch into one of the tumor subtypes or non-tumor. We call this problem “learning from partial label proportions (LPLP)” and formulate the problem as a weakly supervised learning problem. Then, we propose an efficient algorithm for this challenging problem by decomposing it into two weakly supervised learning subproblems: multiple instance learning (MIL) and learning from label proportions (LLP). These subproblems are optimized efficiently in an end-to-end manner. The effectiveness of our algorithm is demonstrated through experiments conducted on two WSI datasets. This code is available at https://github.com/matsuo-shinnosuke/LPLP.

DOI： 10.1007/978-3-031-72120-5_35

Web of Science

Scopus

CiNii Research

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Modern Pathology  

Neoadjuvant therapies are used for locally advanced non–small cell lung carcinomas, whereby pathologists histologically evaluate the effect using resected specimens. Major pathological response (MPR) has recently been used for treatment evaluation and as an economical survival surrogate; however, interobserver variability and poor reproducibility are often noted. The aim of this study was to develop a deep learning (DL) model to predict MPR from hematoxylin and eosin–stained tissue images and to validate its utility for clinical use. We collected data on 125 primary non–small cell lung carcinoma cases that were resected after neoadjuvant therapy. The cases were randomly divided into 55 for training/validation and 70 for testing. A total of 261 hematoxylin and eosin–stained slides were obtained from the maximum tumor beds, and whole slide images were prepared. We used a multiscale patch model that can adaptively weight multiple convolutional neural networks trained with different field-of-view images. We performed 3-fold cross-validation to evaluate the model. During testing, we compared the percentages of viable tumor evaluated by annotator pathologists (reviewed data), those evaluated by nonannotator pathologists (primary data), and those predicted by the DL-based model using 2-class confusion matrices and receiver operating characteristic curves and performed a survival analysis between MPR-achieved and non-MPR cases. In cross-validation, accuracy and mean F1 score were 0.859 and 0.805, respectively. During testing, accuracy and mean F1 score with reviewed data and those with primary data were 0.986, 0.985, 0.943, and 0.943, respectively. The areas under the receiver operating characteristic curve with reviewed and primary data were 0.999 and 0.978, respectively. The disease-free survival of MPR-achieved cases with reviewed and primary data was significantly better than that of the non-MPR cases (P<.001 and P=.001), and that predicted by the DL-based model was almost identical (P=.005). The DL model may support pathologist evaluations and can offer accurate determinations of MPR in patients.

DOI： 10.1016/j.modpat.2023.100302

Web of Science

Scopus

PubMed

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Proceedings of the IEEE International Conference on Computer Vision  

Learning from label proportions (LLP) is a promising weakly supervised learning problem. In LLP, a set of instances (bag) has label proportions, but no instance-level labels are given. LLP aims to train an instance-level classifier by using the label proportions of the bag. In this paper, we propose a bag-level data augmentation method for LLP called MixBag, based on the key observation from our preliminary experiments; that the instance-level classification accuracy improves as the number of labeled bags increases even though the total number of instances is fixed. We also propose a confidence interval loss designed based on statistical theory to use the augmented bags effectively. To the best of our knowledge, this is the first attempt to propose bag-level data augmentation for LLP. The advantage of MixBag is that it can be applied to instance-level data augmentation techniques and any LLP method that uses the proportion loss. Experimental results demonstrate this advantage and the effectiveness of our method.

DOI： 10.1109/ICCV51070.2023.01519

Web of Science

Scopus

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

Detection of mitosis events plays an important role in biomedical research. Deep-learning-based mitosis detection methods have achieved outstanding performance with a certain amount of labeled data. However, these methods require annotations for each imaging condition. Collecting labeled data involves time-consuming human labor. In this paper, we propose a mitosis detection method that can be trained with partially annotated sequences. The base idea is to generate a fully labeled dataset from the partial labels and train a mitosis detection model with the generated dataset. First, we generate an image pair not containing mitosis events by frame-order flipping. Then, we paste mitosis events to the image pair by alpha-blending pasting and generate a fully labeled dataset. We demonstrate the performance of our method on four datasets, and we confirm that our method outperforms other comparisons which use partially labeled sequences. Code is available at https://github.com/naivete5656/MDPAFOF.

DOI： 10.1007/978-3-031-43993-3_47

Web of Science

Scopus

Language：English   Publishing type：Research paper (international conference proceedings)  

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Digestive Endoscopy  

Objectives: Existing endoscopic scores for ulcerative colitis (UC) objectively categorize disease severity based on the presence or absence of endoscopic findings; therefore, it may not reflect the range of clinical severity within each category. However, inflammatory bowel disease (IBD) expert endoscopists categorize the severity and diagnose the overall impression of the degree of inflammation. This study aimed to develop an artificial intelligence (AI) system that can accurately represent the assessment of the endoscopic severity of UC by IBD expert endoscopists. Methods: A ranking-convolutional neural network (ranking-CNN) was trained using comparative information on the UC severity of 13,826 pairs of endoscopic images created by IBD expert endoscopists. Using the trained ranking-CNN, the UC Endoscopic Gradation Scale (UCEGS) was used to express severity. Correlation coefficients were calculated to ensure that there were no inconsistencies in assessments of severity made using UCEGS diagnosed by the AI and the Mayo Endoscopic Subscore, and the correlation coefficients of the mean for test images assessed using UCEGS by four IBD expert endoscopists and the AI. Results: Spearman's correlation coefficient between the UCEGS diagnosed by AI and Mayo Endoscopic Subscore was approximately 0.89. The correlation coefficients between IBD expert endoscopists and the AI of the evaluation results were all higher than 0.95 (P < 0.01). Conclusions: The AI developed here can diagnose UC severity endoscopically similar to IBD expert endoscopists.

DOI： 10.1111/den.14677

Web of Science

Scopus

PubMed

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS  

3D cell tracking in a living organism has a crucial role in live cell image analysis. Cell tracking in C. elegans has two difficulties. First, cell migration in a consecutive frame is large since they move their head during scanning. Second, cell detection is often inconsistent in consecutive frames due to touching cells and low-contrast images, and these inconsistent detections affect the tracking performance worse. In this paper, we propose a cell tracking method to address these issues, which has two main contributions. First, we introduce cell position heatmap-based non-rigid alignment with test-time fine-tuning, which can warp the detected points to near the positions at the next frame. Second, we propose a pairwise detection method, which uses the information of detection results at the previous frame for detecting cells at the current frame. The experimental results demonstrate the effectiveness of each module, and the proposed method achieved the best performance in comparison.

DOI： 10.1109/EMBC40787.2023.10340619

Web of Science

Scopus

PubMed

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Experimental Dermatology  

Skin is composed of different layers, including the stratum corneum, epidermal living layer and papillary and reticular dermis. Each has specific optical properties due to differences in their biological components. Alterations in the skin's cutaneous biological components resulting from photoaging caused by chronic exposure to UV light affect the deterioration of appearance associated with the skin's optical properties. Various methods for analysing cutaneous optical properties have been previously proposed, including mathematical models and computer simulations. However, these were insufficient to elucidate changes in each skin layer and comprehensively understand the skin's integrated optical properties. We focused on UV-induced yellowing of the facial skin. We evaluated site-specific optical absorption of human skin tissue sections to investigate the yellowish discoloration, which is suggested to be related to the photodamage process. The method includes our original technique of separating the transmitted and scattered light using high-frequency illumination microscopy, leading to microscopic analysis of the tissue's optical absorption in the regions of interest. In analysing the sun-exposed facial skin tissue sections, we successfully showed that dermal regions of aged skin have increased absorption at 450 nm, where yellowish colours are complemented. Furthermore, we confirmed that elastic fibres with observable histological disorder resulting from photodamage are a prominent source of high optical absorption. We detected changes in the skin's optical absorption associated with dermal degeneration resulting from photodamage using a novel optical microscopy technique. The results provide a base for the evaluation of optical property changes for both yellowing discoloration and other tissue disorders.

DOI： 10.1111/exd.14840

Web of Science

Scopus

PubMed

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings  

This paper proposes a novel and efficient method for Learning from Label Proportions (LLP), whose goal is to train a classifier only by using the class label proportions of instance sets, called bags. We propose a novel LLP method based on an online pseudo-labeling method with regret minimization. As opposed to the previous LLP methods, the proposed method effectively works even if the bag sizes are large. We demonstrate the effectiveness of the proposed method using some benchmark datasets.

DOI： 10.1109/ICASSP49357.2023.10097069

Scopus

Language：English   Publishing type：Research paper (international conference proceedings)  

Language：English   Publishing type：Research paper (international conference proceedings)  

Language：English   Publishing type：Research paper (international conference proceedings)  

Language：English   Publishing type：Research paper (international conference proceedings)  

Web of Science

Web of Science

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Proceedings - 2023 IEEE Winter Conference on Applications of Computer Vision, WACV 2023  

Drone crowd tracking has various applications such as crowd management and video surveillance. Unlike in general multi-object tracking, the size of the objects to be tracked are small, and the ground truth is given by a point-level annotation, which has no region information. This causes the lack of discriminative features for finding the same objects from many similar objects. Thus, similarity-based tracking techniques, which are widely used for multi-object tracking with bounding-box, are difficult to use. To deal with this problem, we take into account the temporal context of the local area. To aggregate temporal context in a local area, we propose a multi-frame attention with feature-level warping. The feature-level warping can align the features of the same object in multiple frames, and then multi-frame attention can effectively aggregate the temporal context from the warped features. The experimental results show the effectiveness of our method. Our method outperformed the state-of-the-art method in DroneCrowd dataset. The code is publicly available in https://github.com/asanomitakanori/mfa-feature-warping.

DOI： 10.1109/WACV56688.2023.00171

Web of Science

Scopus

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Proceedings - 2023 IEEE Winter Conference on Applications of Computer Vision, WACV 2023  

Cell instance segmentation that recognizes each cell boundary is an important task in cell image analysis. While deep learning-based methods have shown promising performances with a certain amount of training data, most of them require full annotations that show the boundary of each cell. Generating the annotation for cell segmentation is time-consuming and human labor. To reduce the annotation cost, we propose a weakly supervised segmentation method using two types of weak labels (one for cell type and one for nuclei position). Unlike general images, these two labels are easily obtained in phase-contrast images. The intercellular boundary, which is necessary for cell instance segmentation, cannot be directly obtained from these two weak labels, so to generate the boundary information, we propose a single instance pasting based on the copy-and-paste technique. First, we locate single-cell regions by counting cells and store them in a pool. Then, we generate the intercel-lular boundary by pasting the stored single-cell regions to the original image. Finally, we train a boundary estimation network with the generated labels and perform instance segmentation with the network. Our evaluation on a public dataset demonstrated that the proposed method achieves the best performance among the several weakly supervised methods we compared.

DOI： 10.1109/WACV56688.2023.00320

Web of Science

Scopus

CiNii Research

Publisher：Proceedings - International Symposium on Biomedical Imaging  

This paper proposes a semi-supervised pathological image segmentation method that can improve the segmentation using a large amount of unlabeled data. Typical pseudo labeling methods select pseudo labels from unlabeled data to be used for re-training based on the confidence of each patch. However, the initial estimation is not accurate, so the pseudo labels contain many inaccurate labels. This may affect the segmentation performance. The highlight of this study is that we avoid selecting incorrect pseudo labels by introducing a spatial distribution model in a whole slide image. This is based on the assumption that a tumor region forms a cluster in tissue. This improves pseudo label selection and segmentation performance. Experimental results demonstrate the effectiveness of our method, where our method achieved the best segmentation performance in comparison.

DOI： 10.1109/ISBI53787.2023.10230407

Web of Science

Scopus

CiNii Research

Publisher：Proceedings - International Symposium on Biomedical Imaging  

The development of medical image segmentation using deep learning can significantly support doctors' diagnoses. Deep learning needs large amounts of data for training, which also requires data augmentation to extend diversity for preventing overfitting. However, the existing methods for data augmentation of medical image segmentation are mainly based on models which need to update parameters and cost extra computing resources. We proposed data augmentation methods designed to train a high-accuracy deep learning network for medical image segmentation. The proposed data augmentation approaches are called KeepMask and KeepMix, which can create medical images by better identifying the boundary of the organ with no more parameters. Our methods achieved better performance and obtained more precise boundaries for medical image segmentation on datasets. The dice coefficient of our methods achieved 94.15% (3.04% higher than baseline) on CHAOS and 74.70% (5.25% higher than baseline) on MSD spleen with Unet.

DOI： 10.1109/ISBI53787.2023.10230495

Web of Science

Scopus

CiNii Research

CiNii Research

Publisher：Proceedings - International Symposium on Biomedical Imaging  

Semi-supervised domain adaptation is a technique to build a classifier for a target domain by modifying a classifier in another (source) domain using many unlabeled samples and a small number of labeled samples from the target domain. In this paper, we develop a semi-supervised domain adaptation method, which has robustness to class-imbalanced situations, which are common in medical image classification tasks. For robustness, we propose a weakly-supervised clustering pipeline to obtain high-purity clusters and utilize the clusters in representation learning for domain adaptation. The proposed method showed state-of-the-art performance in the experiment using severely class-imbalanced pathological image patches.

DOI： 10.1109/ISBI53787.2023.10230451

Web of Science

Scopus

CiNii Research

Publisher：Proceedings - International Symposium on Biomedical Imaging  

The domain shift in pathological segmentation is an important problem, where a network trained by a source domain (collected at a specific hospital) does not work well in the target domain (from different hospitals) due to the different image features. Due to the problems of class imbalance and different class prior of pathology, typical unsupervised domain adaptation methods do not work well by aligning the distribution of source domain and target domain. In this paper, we propose a cluster entropy for selecting an effective whole slide image (WSI) that is used for semi-supervised domain adaptation. This approach can measure how the image features of the WSI cover the entire distribution of the target domain by calculating the entropy of each cluster and can significantly improve the performance of domain adaptation. Our approach achieved competitive results against the prior arts on datasets collected from two hospitals.

DOI： 10.1109/ISBI53787.2023.10230359

Web of Science

Scopus

CiNii Research

Language：English   Publisher：Arthritis Research and Therapy  

Background: X-ray images are commonly used to assess the bone destruction of rheumatoid arthritis. The purpose of this study is to propose an automatic-bone-destruction-evaluation system fully utilizing deep neural networks (DNN). This system detects all target joints of the modified Sharp/van der Heijde score (SHS) from a hand X-ray image. It then classifies every target joint as intact (SHS = 0) or non-intact (SHS ≥ 1). Methods: We used 226 hand X-ray images of 40 rheumatoid arthritis patients. As for detection, we used a DNN model called DeepLabCut. As for classification, we built four classification models that classify the detected joint as intact or non-intact. The first model classifies each joint independently, whereas the second model does it while comparing the same contralateral joint. The third model compares the same joint group (e.g., the proximal interphalangeal joints) of one hand and the fourth model compares the same joint group of both hands. We evaluated DeepLabCut’s detection performance and classification models’ performances. The classification models’ performances were compared to three orthopedic surgeons. Results: Detection rates for all the target joints were 98.0% and 97.3% for erosion and joint space narrowing (JSN). Among the four classification models, the model that compares the same contralateral joint showed the best F-measure (0.70, 0.81) and area under the curve of the precision-recall curve (PR-AUC) (0.73, 0.85) regarding erosion and JSN. As for erosion, the F-measure and PR-AUC of this model were better than the best of the orthopedic surgeons. Conclusions: The proposed system was useful. All the target joints were detected with high accuracy. The classification model that compared the same contralateral joint showed better performance than the orthopedic surgeons regarding erosion.

DOI： 10.1186/s13075-022-02914-7

Web of Science

Scopus

PubMed

Language：English   Publishing type：Research paper (scientific journal)  

DOI： https://doi.org/10.1186/s13075-022-02914-7

Language：English   Publishing type：Research paper (international conference proceedings)  

Language：English   Publishing type：Research paper (international conference proceedings)  

Language：English   Publishing type：Research paper (international conference proceedings)  

Language：English   Publishing type：Research paper (scientific journal)   Publisher：IEEE Access  

For automatic disease-severity-level estimation, a large-scale medical image dataset with level annotations is generally necessary. However, attaching absolute-level annotations (such as levels 0, 1, and 3) is very costly and even inaccurate due to the level ambiguity. In this study, we proved experimentally that using a ranking function for level estimation can relax this difficulty. We propose a multi-task learning method for automatically estimating disease-severity levels that combine learning to rank with regression. The ranking function of the proposed method is trainable by relative-level and a small number of absolute-level annotations. For relative-level annotation, an annotator only needs to specify that one image has a higher disease level than another - this is much easier than absolute-level annotation. The proposed method enables disease-severity classification by calibrating the ranking function based on relative-level annotation through regression. The effectiveness of the method was proved through a large-scale experiment of ulcerative colitis-severity estimation with colonoscopy images.

DOI： 10.1109/ACCESS.2022.3155769

Web of Science

Scopus

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022  

Multi-object tracking is essential in biomedical image analysis. Most methods follow a tracking-by-detection approach that involves using object detectors and learning the appearance feature models of the detected regions for association. Although these methods can learn the appearance similarity features to identify the same objects among frames, they have difficulties identifying the same cells because cells have a similar appearance and their shapes change as they migrate. In addition, cells often partially overlap for several frames. In this case, even an expert biologist would require knowledge of the spatial-temporal context in order to identify individual cells. To tackle such difficult situations, we propose a cell-tracking method that can effectively use the spatial-temporal context in multiple frames by using long-term motion estimation and an object-level warping loss. We conducted experiments showing that the proposed method outperformed state-of-the-art methods under various conditions on real biological images.

DOI： 10.1109/WACV51458.2022.00182

Web of Science

Scopus

Publisher：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

Automatic image-based disease severity estimation generally uses discrete (i.e., quantized) severity labels. Annotating discrete labels is often difficult due to the images with ambiguous severity. An easier alternative is to use relative annotation, which compares the severity level between image pairs. By using a learning-to-rank framework with relative annotation, we can train a neural network that estimates rank scores that are relative to severity levels. However, the relative annotation for all possible pairs is prohibitive, and therefore, appropriate sample pair selection is mandatory. This paper proposes a deep Bayesian active-learning-to-rank, which trains a Bayesian convolutional neural network while automatically selecting appropriate pairs for relative annotation. We confirmed the efficiency of the proposed method through experiments on endoscopic images of ulcerative colitis. In addition, we confirmed that our method is useful even with the severe class imbalance because of its ability to select samples from minor classes automatically.

DOI： 10.1007/978-3-031-12053-4_45

Web of Science

Scopus

Language：Japanese   Publisher：Japan Society of Civil Engineers  

<p>There has been a growing demand to strengthen existing disaster prevention education tobe prepared for the huge tsunami expected to occur in the near future. Virtual reality, which allows people to virtually experience natural disasters, has a strong potential in fostering disaster awareness among citizens. However, it requires enormous human and time resources to map the texture of structures to urban area-imitating virtual space. On the other hand, pix2pixHD proposed by Wang et al. can generate high-resolution synthetic images by learning from reference images, label data, and object boundary data. In this study, we applied pix2pixHD and transfer learning, which diverts networks trained on other similar domains, to verify texture mapping of urban areas in Japan from a limited set of image data.</p>

DOI： 10.11532/jsceiii.3.j2_551

CiNii Research

Language：Japanese   Publisher：MEDICAL IMAGING AND INFORMATION SCIENCES  

<p>Phoacouastic (PA) imaging has been gaining attention as a new imaging modality that can non-invasively visualize blood vessels inside biological tissues. In this paper, the reconstruction process of PA, and its technical issues. In the process of imaging large body parts through multi-scan fusion, alignment turns out to be an important issue, since body motion degrades image quality. In this paper, we carefully examine the characteristics of PA images and propose a novel registration method that achieves better alignment while effectively decomposing the shot volumes into low-rank foreground (blood vessels), dense background (noise), and sparse complement (corruption) components on the basis of the PA characteristics. The results of experiments using a challenging real dataset demonstrate the efficacy of the proposed method, which significantly improved image quality, and had the best alignment accuracy among the state-of-the-art methods tested. We also introduce several clinical application of photoacoustic imaging, such as planning of flap surgery.</p>

DOI： 10.11318/mii.39.14

CiNii Research

Publisher：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

We propose a deep low-rank alignment network that can simultaneously perform non-rigid alignment and noise decomposition for multiple images despite severe noise and sparse corruptions. To address this challenging task, we introduce a low-rank loss in deep learning under the assumption that a set of well-aligned, well-denoised images should be linearly correlated, and thus, that a matrix consisting of the images should be low-rank. This allows us to remove the noise and corruption from input images in a self-supervised learning manner (i.e., without requiring supervised data). In addition, we introduce multi-input attention modules into Siamese U-nets in order to aggregate the corruption information from the set of images. To the best of our knowledge, this is the first attempt to introduce a low-rank loss for deep learning-based non-rigid alignment. Experiments using both synthetic data and real medical image data demonstrate the effectiveness of the proposed method. The code will be publicly available in https://github.com/asanomitakanori/Unsupervised-Deep-Non-Rigid-Alignment-by-Low-Rank-Loss-and-Multi-Input-Attention.

DOI： 10.1007/978-3-031-16446-0_18

Web of Science

Scopus

Publisher：IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops  

Multi-class cell detection (cancer or non-cancer) from a whole slide image (WSI) is an important task for pathological diagnosis. Cancer and non-cancer cells often have a similar appearance, so it is difficult even for experts to classify a cell from a patch image of individual cells. They usually identify the cell type not only on the basis of the appearance of a single cell but also on the context of the surrounding cells. For using such information, we propose a multi-class cell-detection method that introduces a modified self-attention to aggregate the surrounding image features of both classes. Experimental results demonstrate the effectiveness of the proposed method; our method achieved the best performance compared with a method, which simply uses the standard self-attention method.

DOI： 10.1109/CVPRW56347.2022.00202

Web of Science

Scopus

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/EMBC46164.2021.9630809

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/EMBC46164.2021.9630720

Language：English   Publishing type：Research paper (international conference proceedings)  

The domain shift problem is an important issue in automatic cell detection. A detection network trained with training data under a specific condition (source domain) may not work well in data under other conditions (target domain). We propose an unsupervised domain adaptation method for cell detection using the pseudo-cell-position heatmap, where a cell centroid becomes a peak with a Gaussian distribution in the map. In the prediction result for the target domain, even if a peak location is correct, the signal distribution around the peak often has a non-Gaussian shape. The pseudo-cell-position heatmap is re-generated using the peak positions in the predicted heatmap to have a clear Gaussian shape. Our method selects confident pseudo-cell-position heatmaps using a Bayesian network and adds them to the training data in the next iteration. The method can incrementally extend the domain from the source domain to the target domain in a semi-supervised manner. In the experiments using 8 combinations of domains, the proposed method outperformed the existing domain adaptation methods.

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： https://doi.org/10.1007/978-3-030-87196-3_44

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1007/978-3-030-87237-3_36

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1007/978-3-030-87237-3_41

Language：English   Publishing type：Research paper (international conference proceedings)  

Language：English   Publishing type：Research paper (scientific journal)  

Language：English   Publishing type：Research paper (international conference proceedings)  

Automated mitotic detection in time-lapse phase-contrast microscopy provides us much information for cell behavior analysis, and thus several mitosis detection methods have been proposed. However, these methods still have two problems; 1) they cannot detect multiple mitosis events when there are closely placed. 2) they do not consider the annotation gaps, which may occur since the appearances of mitosis cells are very similar before and after the annotated frame. In this paper, we propose a novel mitosis detection method that can detect multiple mitosis events in a candidate sequence and mitigate the human annotation gap via estimating a spatiotemporal likelihood map by 3DCNN. In this training, the loss gradually decreases with the gap size between ground truth and estimation. This mitigates the annotation gaps. Our method outperformed the compared methods in terms of F1- score using a challenging dataset that contains the data under four different conditions.

Language：English   Publishing type：Research paper (scientific journal)  

The geometric organization of collagen fibers in human reticular dermis and its relationship to that of elastic fibers remain unclear. The tight packing and complex intertwining of dermal collagen fibers hinder accurate analysis of fiber orientation. We hypothesized that combined multiphoton microscopy and biaxial extension could overcome this issue. Continuous observation of fresh dermal sheets under biaxial extension revealed that the geometry of the elastic fiber network is maintained during expansion. Full-thickness human thigh skin samples were biaxially extended and cleared to visualize the entire reticular dermis. Throughout the dermis, collagen fibers straightened with increased inter-fiber spaces, making them more clearly identifiable after extension. The distribution of collagen fibers was evaluated with compilation of local orientation data. Two or three modes were confirmed in all superficial reticular layer samples. A high degree of local similarities in the direction of collagen and elastic fibers was observed. More than 80% of fibers had directional differences of ≤15°, regardless of layer. Understanding the geometric organization of fibers in the reticular dermis improves the understanding of mechanisms underlying the pliability of human skin. Combined multiphoton imaging and biaxial extension provides a research tool for studying the fibrous microarchitecture of the skin.

DOI： 10.1038/s41598-019-47213-5

Language：English   Publishing type：Research paper (international conference proceedings)  

Cell behavior analysis in high-throughput biological experiments is important for research and discovery in biology and medicine. To perform the high-throughput experiments, it requires to capture images in low frame rate in order to record images on multi-points. In such a low frame rate image sequence, movements of cells between successive frames are often larger than distances to nearby cells, and thus current methods based on proximity do not work properly. In this study, we propose a cell tracking method that enables to track cells in low frame rate by simultaneously estimating all of the cell motions in successive frames. In the experiments under dense conditions in low frame rate, our method outperformed the other methods.

DOI： 10.1007/978-3-030-32239-7_44

Language：English   Publishing type：Research paper (international conference proceedings)  

We propose a soft-constrained clustering method for group-based labeling of medical images. Since the idea of group-based labeling is to attach the label to a group of samples at once, we need to have groups (i.e., clusters) with high purity. The proposed method is formulated to achieve high purity even for difficult clustering tasks such as medical image clustering, where image samples of the same class are often very distant in their feature space. In fact, those images degrade the performance of conventional constrained clustering methods. Experiments with an endoscopy image dataset demonstrated that our method outperformed various state-of-the-art methods.

DOI： 10.1007/978-3-030-32254-0_47

Language：English   Publishing type：Research paper (international conference proceedings)  

Cell shape analysis is important in biomedical research. Deep learning methods may perform to segment individual cells if they use sufficient training data that the boundary of each cell is annotated. However, it is very time-consuming for preparing such detailed annotation for many cell culture conditions. In this paper, we propose a weakly supervised method that can segment individual cell regions who touch each other with unclear boundaries in dense conditions without the training data for cell regions. We demonstrated the efficacy of our method using several data-set including multiple cell types captured by several types of microscopy. Our method achieved the highest accuracy compared with several conventional methods. In addition, we demonstrated that our method can perform without any annotation by using fluorescence images that cell nuclear were stained as training data. Code is publicly available in https://github.com/naivete5656/WSISPDR.

DOI： 10.1007/978-3-030-32239-7_72

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/EMBC.2019.8857011

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/EMBC.2019.8856465

Language：English   Publishing type：Research paper (international conference proceedings)  

Photoacoustic (PA) imaging is a promising new imaging technology for non-invasively visualizing blood vessels inside biological tissues. In addition to blood vessels, body hairs are also visualized in PA imaging, and the body hair signals degrade the visibility of blood vessels. For learning a body hair classifier, the amount of real training and test data is limited, because PA imaging is a new modality. To address this problem, we propose a novel semi-supervised learning (SSL) method for extracting body hairs. The method effectively learns the discriminative model from small labeled training data and small unlabeled test data by introducing prior knowledge, of the orientation similarity among adjacent body hairs, into SSL. Experimental results using real PA data demonstrate that the proposed approach is effective for extracting body hairs as compared with several baseline methods.

DOI： 10.1109/ISBI.2019.8759249

Language：English   Publishing type：Research paper (scientific journal)  

This study aimed to characterize in vivo human digital arteries in three-dimensions using photoacoustic tomography in order to understand the specific mechanism underlying arterial deformation associated with movement of the proximal interphalangeal joint. Three-dimensional morphological data were obtained on the radialis indicis artery (radial artery of the index finger) at different angles of the joint. The association between increased curvature of the deformation and the anatomical region was assessed. Characteristic morphological deformations in areas of major deformation were determined. The deformation of the artery was characterized by three consecutive curves in juxta-articular regions, which were particularly noticeable when the joint was flexed at an angle of ≥ 60°. The change in the curvature of the deformation during 30°–90° of flexion was lower in middle-aged individuals than in young individuals. Better understanding of the mechanism underlying deformation of the digital arteries may contribute to advancements in flap procedures and rehabilitation strategies after digital artery repair.

DOI： 10.1177/1753193418807833

Language：English   Publishing type：Research paper (scientific journal)   Publisher：IEEE Transactions on Pattern Analysis and Machine Intelligence  

Recognizing wet surfaces and their degrees of wetness is essential for many computer vision applications. Surface wetness can inform us slippery spots on a road to autonomous vehicles, muddy areas of a trail to humanoid robots, and the freshness of groceries to us. The fact that surfaces darken when wet, i.e., monochromatic appearance change, has been modeled to recognize wet surfaces in the past. In this paper, we show that color change, particularly in its spectral behavior, carries rich information about surface wetness. We first derive an analytical spectral appearance model of wet surfaces that expresses the characteristic spectral sharpening due to multiple scattering and absorption in the surface. We present a novel method for estimating key parameters of this spectral appearance model, which enables the recovery of the original surface color and the degree of wetness from a single multispectral image. Applied to a multispectral image, the method estimates the spatial map of wetness together with the dry spectral distribution of the surface. To our knowledge, this is the first work to model and leverage the spectral characteristics of wet surfaces to decipher its appearance. We conduct comprehensive experimental validation with a number of wet real surfaces. The results demonstrate the accuracy of our model and the effectiveness of our method for surface wetness and color estimation.

DOI： 10.1109/TPAMI.2019.2903496

Scopus

Other Link： https://ieeexplore.ieee.org/document/8661549

Language：English   Publishing type：Research paper (scientific journal)  

Phase contrast time-lapse microscopy is a non-destructive technique that generates large volumes of image-based information to quantify the behaviour of individual cells or cell populations. To guide the development of algorithms for computer-aided cell tracking and analysis, 48 time-lapse image sequences, each spanning approximately 3.5 days, were generated with accompanying ground truths for C2C12 myoblast cells cultured under 4 different media conditions, including with fibroblast growth factor 2 (FGF2), bone morphogenetic protein 2 (BMP2), FGF2 + BMP2, and control (no growth factor). The ground truths generated contain information for tracking at least 3 parent cells and their descendants within these datasets and were validated using a two-tier system of manual curation. This comprehensive, validated dataset will be useful in advancing the development of computer-aided cell tracking algorithms and function as a benchmark, providing an invaluable opportunity to deepen our understanding of individual and population-based cell dynamics for biomedical research.

DOI： 10.1038/sdata.2018.237

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1016/j.jdermsci.2018.07.006

Language：English   Publishing type：Research paper (international conference proceedings)  

In transmitted light microscopy, a specimen tends to be observed as unclear. This is caused by a phenomenon that an image sensor captures the sum of these scattered light rays traveled from different paths due to scattering. To cope with this problem, we propose a novel computational photography approach for separating directly transmitted light from the scattering light in a transmitted light microscope by using high-frequency lighting. We first investigated light paths and clarified what types of light overlap in transmitted light microscopy. The scattered light can be simply represented and removed by using the difference in observations between focused and unfocused conditions, where the high-frequency illumination becomes homogeneous. Our method makes a novel spatial multiple-spectral absorption analysis possible, which requires absorption coefficients to be measured in each spectrum at each position. Experiments on real biological tissues demonstrated the effectiveness of our method.

DOI： 10.1007/978-3-319-66185-8_2

Language：English   Publishing type：Research paper (international conference proceedings)  

We propose a fully automatic system to reconstruct and visualize 3D blood vessels in Augmented Reality (AR) system from stereo X-ray images with bones and body fat. Currently, typical 3D imaging technologies are expensive and carrying the risk of irradiation exposure. To reduce the potential harm, we only need to take two X-ray images before visualizing the vessels. Our system can effectively reconstruct and visualize vessels in following steps. We first conduct initial segmentation using Markov Random Field and then refine segmentation in an entropy based post-process. We parse the segmented vessels by extracting their centerlines and generating trees. We propose a coarse-to-fine scheme for stereo matching, including initial matching using affine transform and dense matching using Hungarian algorithm guided by Gaussian regression. Finally, we render and visualize the reconstructed model in a HoloLens based AR system, which can essentially change the way of visualizing medical data. We have evaluated its performance by using synthetic and real stereo X-ray images, and achieved satisfactory quantitative and qualitative results.

DOI： 10.1109/ICCVW.2017.20

Language：English   Publishing type：Research paper (international conference proceedings)  

In this paper, we focus on semi-supervised learning for biomedical image segmentation, so as to take advantage of huge unlabelled data. We observe that there usually exist some homogeneous connected areas of low confidence in biomedical images, which tend to confuse the classifier trained with limited labelled samples. To cope with this difficulty, we propose to construct forest oriented super pixels(voxels) to augment the standard random forest classifier, in which super pixels(voxels) are built upon the forest based code. Compared to the state-of-the-art, our proposed method shows superior segmentation performance on challenging 2D/3D biomedical images. The full implementation (based on Matlab) is available at https://github.com/lingucv/ssl_superpixels.

DOI： 10.1007/978-3-319-66182-7_80

Language：English   Publishing type：Research paper (international conference proceedings)  

Photoacoustic (PA) imaging has been gaining attention as a new imaging modality that can non-invasively visualize blood vessels inside biological tissues. In the process of imaging large body parts through multi-scan fusion,alignment turns out to be an important issue,since body motion degrades image quality. In this paper,we carefully examine the characteristics of PA images and propose a novel registration method that achieves better alignment while effectively decomposing the shot volumes into low-rank foreground (blood vessels),dense background (noise),and sparse complement (corruption) components on the basis of the PA characteristics. The results of experiments using a challenging real data-set demonstrate the efficacy of the proposed method,which significantly improved image quality,and had the best alignment accuracy among the state-of-the-art methods tested.

DOI： 10.1007/978-3-319-46726-9_38

Language：English   Publishing type：Research paper (international conference proceedings)  

A newly developed imaging technique called light-sheet laser microscopy imaging can visualize the detailed 3D structures of capillaries. Capillaries form complicated network structures in the obtained data, and this makes it difficult to model vessel structures by existing methods that implicitly assume simple tree structures for blood vessels. To cope with such dense capillaries with network structures, we propose to track the flow of blood vessels along a base-axis using a multiple-object tracking framework. We first track multiple blood vessels in cross-sectional images along a single axis to make the trajectories of blood vessels, and then connect these blood vessels to reveal their entire structures. This framework is efficient to track densely distributed vessels since it uses only a single cross-sectional plane. The network structure is then generated in the post-processing by connecting blood vessels on the basis of orientations of the trajectories. The results of experiments using a challenging real data-set demonstrate the efficacy of the proposed method, which are capable of modeling dense capillaries.

DOI： 10.1109/CVPRW.2016.168

Language：English   Publishing type：Research paper (scientific journal)  

The in vitro formation of network structures derived from endothelial cells in grafts before transplantation contributes to earlier engraftment. In a previous study, endothelial cells migrated to form a net-shaped structure in co-culture. However, the specific network formation behavior of endothelial cells during migration remains unclear. In this study, we demonstrated the tracing behavior and cell cycle of endothelial cells using Fucci-labeled (Fluorescent Ubiquitination-based Cell Cycle Indicator) endothelial cells. Here, we observed the co-culture of Fucci-labeled human umbilical vein endothelial cells (HUVECs) together with normal human dermal fibroblasts (NHDFs) using time-lapse imaging and analyzed by multicellular concurrent tracking. In the G0/G1 period, HUVECs migrate faster than in the S/G2/M period, because G0/G1 is the mobile phase and S/G2/M is the proliferation phase in the cell cycle. When HUVECs are co-cultured, they tend to move randomly until they find existing tracks that they then follow to form clusters. Extracellular matrix (ECM) staining showed that collagen IV, laminin and thrombospondin deposited in accordance with endothelial cell networks. Therefore the HUVECs may migrate on the secreted ECM and exhibit tracing behavior, where the HUVECs migrate toward each other. These results suggested that ECM and a cell phase contributed to form a network by accelerating cell migration.

DOI： 10.1016/j.mvr.2015.12.005

Language：English   Publishing type：Research paper (international conference proceedings)  

Automated tracking of cell population is an important element of research and discovery in the biology field. In this paper, we propose a method that tracks cells under highly confluent conditions by using the candidate cell region detection-based association approach. Unlike conventional segmentation-based association tracking methods, the proposed method uses the tracking results from the previous frame to segment the cell regions at the current frame. First, candidate cell regions are detected, and while there may be many false positives, there are very few false negatives. Next, optimized detection results are selected from the candidate regions and associated with the tracking results of the previous frame by resolving a linear programming problem. We quantitatively evaluated the proposed method using a variety of sequences. Results showed that our method has a better tracking performance than conventional segmentation-based association methods.

DOI： 10.2316/P.2013.791-057

Language：English   Publishing type：Research paper (scientific journal)  

Therapies using adult stem cells often require mechanical manipulation such as injection or incorporation into scaffolds. However, force-induced rupture and mechanosensitivity of cells during manipulation is largely ignored. Here, we image cell mechanical structures and perform a biophysical characterization of three different types of human adult stem cells: bone marrow CD34+ hematopoietic, bone marrow mesenchymal and perivascular mesenchymal stem cells. We use micropipette aspiration to characterize cell mechanics and quantify deformation of subcellular structures under force and its contribution to global cell deformation. Our results suggest that CD34+ cells are mechanically suitable for injection systems since cells transition from solid- to fluid-like at constant aspiration pressure, probably due to a poorly developed actin cytoskeleton. Conversely, mesenchymal stem cells from the bone marrow and perivascular niches are more suitable for seeding into biomaterial scaffolds since they are mechanically robust and have developed cytoskeletal structures that may allow cellular stable attachment and motility through solid porous environments. Among these, perivascular stem cells cultured in 6% oxygen show a developed cytoskeleton but a more compliant nucleus, which can facilitate the penetration into pores of tissues or scaffolds. We confirm the relevance of our measurements using cell motility and migration assays and measure survival of injected cells. Since different types of adult stem cells can be used for similar applications, we suggest considering mechanical properties of stem cells to match optimal mechanical characteristics of therapies.

DOI： 10.1016/j.jbiomech.2012.01.032

Language：English   Publishing type：Research paper (scientific journal)  

The wound healing assay in vitro is widely used for research and discovery in biology and medicine. This assay allows for observing the healing process in vitro in which the cells on the edges of the artificial wound migrate toward the wound area. The influence of different culture conditions can be measured by observing the change in the size of the wound area. For further investigation, more detailed measurements of the cell behaviors are required. In this paper, we present an application of automatic cell tracking in phase-contrast microscopy images to wound healing assay. The cell behaviors under three different culture conditions have been analyzed. Our cell tracking system can track individual cells during the healing process and provide detailed spatio-temporal measurements of cell behaviors. The application demonstrates the effectiveness of automatic cell tracking for quantitative and detailed analysis of the cell behaviors in wound healing assay in vitro.

Language：English   Publishing type：Research paper (scientific journal)  

Current cell culture practices are dependent upon human operators and remain laborious and highly subjective, resulting in large variations and inconsistent outcomes, especially when using visual assessments of cell confluency to determine the appropriate time to subculture cells. Although efforts to automate cell culture with robotic systems are underway, the majority of such systems still require human intervention to determine when to subculture. Thus, it is necessary to accurately and objectively determine the appropriate time for cell passaging. Optimal stem cell culturing that maintains cell pluripotency while maximizing cell yields will be especially important for efficient, cost-effective stem cell-based therapies. Toward this goal we developed a real-time computer vision-based system that monitors the degree of cell confluency with a precision of 0.791±0.031 and recall of 0.559±0.043. The system consists of an automated phase-contrast time-lapse microscope and a server. Multiple dishes are sequentially imaged and the data is uploaded to the server that performs computer vision processing, predicts when cells will exceed a pre-defined threshold for optimal cell confluency, and provides a Web-based interface for remote cell culture monitoring. Human operators are also notified via text messaging and e-mail 4 hours prior to reaching this threshold and immediately upon reaching this threshold. This system was successfully used to direct the expansion of a paradigm stem cell population, C2C12 cells. Computer-directed and human-directed control subcultures required 3 serial cultures to achieve the theoretical target cell yield of 50 million C2C12 cells and showed no difference for myogenic and osteogenic differentiation. This automated vision-based system has potential as a tool toward adaptive real-time control of subculturing, cell culture optimization and quality assurance/quality control, and it could be integrated with current and developing robotic cell cultures systems to achieve complete automation.

DOI： 10.1371/journal.pone.0027672

Language：English   Publishing type：Research paper (international conference proceedings)  

Automated cell tracking in populations is important for research and discovery in biology and medicine. In this paper, we propose a cell tracking method based on global spatio-temporal data association which considers hypotheses of initialization, termination, translation, division and false positive in an integrated formulation. Firstly, reliable tracklets (i.e., short trajectories) are generated by linking detection responses based on frame-by-frame association. Next, these tracklets are globally associated over time to obtain final cell trajectories and lineage trees. During global association, tracklets form tree structures where a mother cell divides into two daughter cells. We formulate the global association for tree structures as a maximum-a-posteriori (MAP) problem and solve it by linear programming. This approach is quantitatively evaluated on sequences with thousands of cells captured over several days.

DOI： 10.1109/ISBI.2011.5872571

Language：English   Publishing type：Research paper (international conference proceedings)  

Automated visual-tracking systems of stem cell populations in vitro allow for high-throughput analysis of time-lapse phase-contrast microscopy. In these systems, detection of mitosis, or cell division, is critical to tracking performance as mitosis causes branching of the trajectory of a mother cell into the two trajectories of its daughter cells. Recently, one mitosis detection algorithm showed its success in detecting the time and location that two daughter cells first clearly appear as a result of mitosis. This detection result can therefore helps trajectories to correctly bifurcate and the relations between mother and daughter cells to be revealed. In this paper, we demonstrate that the functionality of this recent mitosis detection algorithm significantly improves state-of-the-art cell tracking systems through extensive experiments on 48 C2C12 myoblastic stem cell populations under four different conditions.

DOI： 10.1109/ISBI.2011.5872832

Language：English   Publishing type：Research paper (scientific journal)  

Due to the enormous potential and impact that stem cells may have on regenerative medicine, there has been a rapidly growing interest for tools to analyze and characterize the behaviors of these cells in vitro in an automated and high throughput fashion. Among these behaviors, mitosis, or cell division, is important since stem cells proliferate and renew themselves through mitosis. However, current automated systems for measuring cell proliferation often require destructive or sacrificial methods of cell manipulation such as cell lysis or in vitro staining. In this paper, we propose an effective approach for automated mitosis detection using phase-contrast time-lapse microscopy, which is a nondestructive imaging modality, thereby allowing continuous monitoring of cells in culture. In our approach, we present a probabilistic model for event detection, which can simultaneously 1) identify spatio-temporal patch sequences that contain a mitotic event and 2) localize a birth event, defined as the time and location at which cell division is completed and two daughter cells are born. Our approach significantly outperforms previous approaches in terms of both detection accuracy and computational efficiency, when applied to multipotent C3H10T1/2 mesenchymal and C2C12 myoblastic stem cell populations.

DOI： 10.1109/TMI.2010.2089384

Language：English   Publishing type：Research paper (international conference proceedings)  

We present several algorithms for cell image analysis including microscopy image restoration, cell event detection and cell tracking in a large population. The algorithms are integrated into an automated system capable of quantifying cell proliferation metrics in vitro in real-time. This offers unique opportunities for biological applications such as efficient cell behavior discovery in response to different cell culturing conditions and adaptive experiment control. We quantitatively evaluated our system's performance on 16 microscopy image sequences with satisfactory accuracy for biologists' need. We have also developed a public website compatible to the system's local user interface, thereby allowing biologists to conveniently check their experiment progress online. The website will serve as a community resource that allows other research groups to upload their cell images for analysis and comparison.

DOI： 10.1109/WACV.2011.5711528

Language：English   Publishing type：Research paper (scientific journal)  

Due to the enormous potential and impact that stem cells may have on regenerative medicine, there has been a rapidly growing interest for tools to analyze and characterize the behaviors of these cells in vitro in an automated and high throughput fashion. Among these behaviors, mitosis, or cell division, is important since stem cells proliferate and renew themselves through mitosis. However, current automated systems for measuring cell proliferation often require destructive or sacrificial methods of cell manipulation such as cell lysis or in vitro staining. In this paper, we propose an effective approach for automated mitosis detection using phase-contrast time-lapse microscopy, which is a nondestructive imaging modality, thereby allowing continuous monitoring of cells in culture. In our approach, we present a probabilistic model for event detection, which can simultaneously 1) identify spatio-temporal patch sequences that contain a mitotic event and 2) localize a birth event, defined as the time and location at which cell division is completed and two daughter cells are born. Our approach significantly outperforms previous approaches in terms of both detection accuracy and computational efficiency, when applied to multipotent C3H10T1/2 mesenchymal and C2C12 myoblastic stem cell populations.

DOI： 10.1109/TMI.2010.2089384

Language：English   Publishing type：Research paper (international conference proceedings)  

We present a method for robustly detecting hematopoietic stem cells (HSCs) in phase contrast microscopy images. HSCs appear to be easy to detect since they typically appear as round objects. However, when HSCs are touching and overlapping, showing the variations in shape and appearance, standard pattern detection methods, such as Hough transform and correlation, do not perform well. The proposed method exploits the output pattern of a ring filter bank applied to the input image, which consists of a series of matched filters with multiple-radius ring-shaped templates. By modeling the profile of each filter response as a quadratic surface, we explore the variations of peak curvatures and peak values of the filter responses when the ring radius varies. The method is validated on thousands of phase contrast microscopy images with different acquisition settings, achieving 96.5% precision and 94.4% recall.

DOI： 10.1109/ISBI.2010.5490394

Language：English   Publishing type：Research paper (international conference proceedings)  

Cell segmentation in microscopy imagery is essential for many bioimage applications such as cell tracking. To segment cells from the background accurately, we present a pixel classification approach that is independent of cell type or imaging modality. We train a set of Bayesian classifiers from clustered local training image patches. Each Bayesian classifier is an expert to make decision in its specific domain. The decision from the mixture of experts determines how likely a new pixel is a cell pixel. We demonstrate the effectiveness of this approach on four cell types with diverse morphologies under different microscopy imaging modalities.

DOI： 10.1109/ISBI.2010.5490399

Language：English   Publishing type：Research paper (international conference proceedings)  

Language：English   Publishing type：Research paper (scientific journal)  

Realization of associative memories by cellular neural networks (CNNs) with binary output is studied. Concerning this problem, a CNN design method based upon generalized eigenvalue minimization (GEVM) has recently been proposed. In this brief, a new CNN design method which is based on the GEVM-based method will be presented. We first give some analytical results related to the basin of attraction of a memory vector. We then derive the design method by combining these analytical results and the GEVM-based method. We finally show through computer simulations that the proposed method can achieve higher recall probability than the original GEVM-based method.

DOI： 10.1109/TCSI.2003.819827

Language：English   Publishing type：Research paper (scientific journal)  

Realization of associative memories by cellular neural networks (CNNs) with binary output is studied. Concerning this problem, a CNN design method based upon generalized eigenvalue minimization (GEVM) has recently been proposed. In this brief, a new CNN design method which is based on the GEVM-based method will be presented. We first give some analytical results related to the basin of attraction of a memory vector. We then derive the design method by combining these analytical results and the GEVM-based method. We finally show through computer simulations that the proposed method can achieve higher recall probability than the original GEVM-based method.

DOI： 10.1109/TCSI.2003.819827

Language：English   Publishing type：Research paper (international conference proceedings)  

Event date： 2024.3

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：沖縄青年会館，那覇市   Country：Japan  

Event date： 2024.3

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：広島大学，東広島市   Country：Japan  

Event date： 2024.3

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：沖縄青年会館，那覇市   Country：Japan  

Event date： 2024.3

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：沖縄青年会館，那覇市   Country：Japan  

Event date： 2023.12

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：富山国際会議場，富山市   Country：Japan  

Event date： 2023.12

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：富山国際会議場，富山市   Country：Japan  

Event date： 2023.12 - 2022.12

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：富山国際会議場，富山市   Country：Japan  

Event date： 2023.11

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：鳥取県立生涯学習センター, 鳥取市   Country：Japan  

Event date： 2023.11

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：鳥取県立生涯学習センター, 鳥取市   Country：Japan  

Event date： 2023.11

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：久留米シティプラザ，久留米市   Country：Japan  

Event date： 2023.10

Language：English   Presentation type：Oral presentation (general)  

Event date： 2023.9

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2023.9

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2023.9

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2023.9

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2023.7

Language：Japanese  

Venue：浜松   Country：Japan  

Event date： 2023.7

Language：Japanese  

Venue：浜松   Country：Japan  

Event date： 2023.7

Language：Japanese  

Venue：浜松   Country：Japan  

Event date： 2023.3

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：沖縄青年会館，那覇市   Country：Japan  

Event date： 2023.3

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：沖縄青年会館，那覇市   Country：Japan  

Event date： 2023.3

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：はこだて未来大学，函館市   Country：Japan  

Event date： 2023.3

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：はこだて未来大学，函館市   Country：Japan  

Event date： 2023.1

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2022.7

Language：Japanese  

Venue：姫路   Country：Japan  

Event date： 2022.7

Language：Japanese  

Venue：姫路   Country：Japan  

Event date： 2022.7

Language：Japanese  

Venue：姫路   Country：Japan  

Event date： 2022.7

Language：Japanese  

Venue：姫路   Country：Japan  

Event date： 2022.7

Language：Japanese  

Venue：姫路   Country：Japan  

Event date： 2022.5

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：豊田工業大学   Country：Japan  

Event date： 2022.2

Language：Japanese   Presentation type：Oral presentation (invited, special)  

Venue：online   Country：Japan  

Event date： 2022.1

Language：Japanese   Presentation type：Oral presentation (invited, special)  

Venue：online   Country：Japan  

Event date： 2021.12

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2021.12

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2021.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2021.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2021.8

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2021.7

Language：Japanese  

Venue：online   Country：Japan  

Event date： 2021.7

Language：Japanese  

Venue：online   Country：Japan  

Event date： 2021.7

Language：Japanese  

Venue：online   Country：Japan  

Event date： 2021.7

Language：Japanese  

Venue：online   Country：Japan  

Event date： 2020.10

Language：English   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2020.10

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2020.10

Language：Japanese  

Country：Japan  

Event date： 2020.10

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2020.10

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2020.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2020.9

Language：English   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2020.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2020.8

Language：Japanese  

Venue：online   Country：Japan  

Event date： 2020.8

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2020.8

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2020.8

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2020.7

Language：Japanese  

Country：Japan  

Event date： 2020.7

Language：Japanese   Presentation type：Oral presentation (invited, special)  

Venue：online   Country：Japan  

Event date： 2020.5

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2020.5

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：online   Country：Japan  

Event date： 2020.3

Language：English   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2020.3

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2020.2

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：静岡   Country：Japan  

Event date： 2020.1

Language：Japanese  

Venue：東京   Country：Japan  

Event date： 2020.1

Language：Japanese  

Venue：東京   Country：Japan  

Event date： 2019.12

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：大分大学、大分   Country：Japan  

Event date： 2019.11

Language：English  

Venue：Shiga   Country：Japan  

Event date： 2019.11

Language：English  

Venue：Shiga   Country：Japan  

Event date： 2019.11

Language：English  

Venue：Shiga   Country：Japan  

Event date： 2019.11

Language：English  

Venue：Shiga   Country：Japan  

Event date： 2019.10

Language：English  

Venue：Shenzhen   Country：China  

We propose a soft-constrained clustering method for group-based labeling of medical images. Since the idea of group-based labeling is to attach the label to a group of samples at once, we need to have groups (i.e., clusters) with high purity. The proposed method is formulated to achieve high purity even for difficult clustering tasks such as medical image clustering, where image samples of the same class are often very distant in their feature space. In fact, those images degrade the performance of conventional constrained clustering methods. Experiments with an endoscopy image dataset demonstrated that our method outperformed various state-of-the-art methods.

Event date： 2019.10

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：東京大学、東京   Country：Japan  

Event date： 2019.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：岡山大学、岡山   Country：Japan  

Event date： 2019.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：九州工業大学、福岡   Country：Japan  

Event date： 2019.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：九州工業大学、福岡   Country：Japan  

Event date： 2019.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：九州工業大学、福岡   Country：Japan  

Event date： 2019.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：岡山大学、岡山   Country：Japan  

Event date： 2019.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：岡山大学、岡山   Country：Japan  

Event date： 2019.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：岡山大学、岡山   Country：Japan  

Event date： 2019.7

Language：English   Presentation type：Oral presentation (general)  

Venue：Berlin   Country：Germany  

Event date： 2019.7

Language：English  

Venue：Berlin   Country：Germany  

Event date： 2019.7 - 2019.8

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：大阪府立国際会議場、大阪   Country：Japan  

Event date： 2019.7 - 2019.8

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：大阪府立国際会議場、大阪   Country：Japan  

Event date： 2019.7 - 2019.8

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：大阪府立国際会議場、大阪   Country：Japan  

Event date： 2019.6

Language：Japanese  

Venue：神戸   Country：Japan  

Event date： 2019.6

Language：Japanese  

Venue：神戸   Country：Japan  

Event date： 2019.5

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：国立オリンピック記念青少年総合センター   Country：Japan  

Event date： 2019.5

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：国立オリンピック記念青少年総合センター   Country：Japan  

Event date： 2019.4

Language：English  

Venue：Venice   Country：Italy  

Photoacoustic (PA) imaging is a promising new imaging technology for non-invasively visualizing blood vessels inside biological tissues. In addition to blood vessels, body hairs are also visualized in PA imaging, and the body hair signals degrade the visibility of blood vessels. For learning a body hair classifier, the amount of real training and test data is limited, because PA imaging is a new modality. To address this problem, we propose a novel semi-supervised learning (SSL) method for extracting body hairs. The method effectively learns the discriminative model from small labeled training data and small unlabeled test data by introducing prior knowledge, of the orientation similarity among adjacent body hairs, into SSL. Experimental results using real PA data demonstrate that the proposed approach is effective for extracting body hairs as compared with several baseline methods.

Event date： 2019.4

Language：Japanese  

Country：Japan  

Event date： 2019.4

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2019.4

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2019.4

Language：English   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2019.4

Language：Japanese  

Venue：USA, San Antonio   Country：Japan  

Event date： 2019.4

Language：English  

Country：United States  

Event date： 2019.4

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2019.4

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2019.4

Language：Japanese  

Country：Japan  

Event date： 2019.4

Language：Japanese  

Country：Japan  

Event date： 2019.4

Language：Japanese  

Country：Japan  

Event date： 2019.4

Language：Japanese  

Country：Japan  

Event date： 2019.4

Language：Japanese  

Country：Japan  

Event date： 2019.4

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2019.4

Language：Japanese  

Country：Japan  

Event date： 2019.3

Language：English  

Country：United States  

Background:
The programmed death receptor 1 (PD-1) and its ligand PD-L1 pathway has shown promising clinical success as an immunotherapy target in cancer treatment, especially in non-small cell lung cancer. Although PD-L1 immunohistochemistry (IHC) is used for the decision on treatment with PD-L1 checkpoint inhibitors, evaluation of PD-L1 IHC may be difficult. Macrophages often exhibit membranous staining and may be misinterpreted as cancer cells. Also, unspecific cytoplasmic staining may occur, and weak partial membranous staining of tumors cells count as positives but may be difficult to detect. The mentioned difficulties are especially troublesome when a very limited proportion of tumor cells such as 1% is sufficient for a positive test, as is the case for some markers, and may lead to significant interobserver variation. Our study purpose was to develop a deep neural network convolution system for automated semi-quantitative assessment of PD-L1 IHC.


Design:
We adopted ensemble of a two-step strategy using the whole slide images of tissue microarray comprising of 343 surgical specimens. Three trained pathologists scored PD-L1 staining proportion into two grades (<50%, >=50%) independently, and discrepant cases were resolved in the discussion with all observers. In the first step, we applied the U-Net architecture to detect all nuclei of tumoral and non-tumoral cells in each core. To create the ground truth labels we annotated bounding boxes for over 4000 nuclei. In the second step, our CNN model classifies the detected cells as either PD-L1 positives (Class 0), negatives (Class 1) or non-tumoral cells (Class 3) (e.g., macrophages, necrotic cells). To train the CNN, pathologists annotated regions so that each region only includes a single class of cells, and the small patch image (64x64) around the detected cells is labeled based on the annotated regions. In the inference step, the trained CNN inferences the class of the detected cells. Based on the inference, the percentage of PD-L1 positive cells of each core was obtained by dividing the number of PD-L1 expressing cancer cells by the total number of cancer cells which is the union of regions represented by Classes 0 and 1.


Results:
Our methods reproduce the expert analysis, leading to an accuracy of 0.96 for automatic diagnosis in the case of total proportion score (TPS) <50%, and 0.92 in the case of TPS >=50%.


Conclusion:
These results clearly open the door to highly-accurate and fully-automatic analysis of PD-L1 IHC.

Other Link： https://www.xcdsystem.com/uscap/program/2019/

Event date： 2019.2

Language：Japanese  

Venue：仙台   Country：Japan  

Event date： 2019.2

Language：Japanese  

Venue：仙台   Country：Japan  

Event date： 2017.10

Language：English  

Venue：Venice   Country：Italy  

We propose a fully automatic system to reconstruct and visualize 3D blood vessels in Augmented Reality (AR) system from stereo X-ray images with bones and body fat. Currently, typical 3D imaging technologies are expensive and carrying the risk of irradiation exposure. To reduce the potential harm, we only need to take two X-ray images before visualizing the vessels. Our system can effectively reconstruct and visualize vessels in following steps. We first conduct initial segmentation using Markov Random Field and then refine segmentation in an entropy based post-process. We parse the segmented vessels by extracting their centerlines and generating trees. We propose a coarse-to-fine scheme for stereo matching, including initial matching using affine transform and dense matching using Hungarian algorithm guided by Gaussian regression. Finally, we render and visualize the reconstructed model in a HoloLens based AR system, which can essentially change the way of visualizing medical data. We have evaluated its performance by using synthetic and real stereo X-ray images, and achieved satisfactory quantitative and qualitative results.

Event date： 2017.10

Language：English   Presentation type：Oral presentation (general)  

Venue：北京大学   Country：Japan  

Event date： 2017.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：琉球大学   Country：Japan  

Event date： 2017.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：琉球大学   Country：Japan  

Event date： 2017.8

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：広島国際会議場   Country：Japan  

Event date： 2017.7

Language：English   Presentation type：Oral presentation (general)  

Venue：Honolulu   Country：United States  

Visual recognition of wet surfaces and their degrees of wetness is important for many computer vision applications. It can inform slippery spots on a road to autonomous vehicles, muddy areas of a trail to humanoid robots, and the freshness of groceries to us. In the past, monochromatic appearance change, the fact that surfaces darken when wet, has been modeled to recognize wet surfaces. In this paper, we show that color change, particularly in its spectral behavior, carries rich information about a wet surface. We derive an analytical spectral appearance model of wet surfaces that expresses the characteristic spectral sharpening due to multiple scattering and absorption in the surface. We derive a novel method for estimating key parameters of this spectral appearance model, which enables the recovery of the original surface color and the degree of wetness from a single observation. Applied to a multispectral image, the method estimates the spatial map of wetness together with the dry spectral distribution of the surface. To our knowledge, this work is the first to model and leverage the spectral characteristics of wet surfaces to revert its appearance. We conduct comprehensive experimental validation with a number of wet real surfaces. The results demonstrate the accuracy of our model and the effectiveness of our method for surface wetness and color estimation.

Event date： 2017.6

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2017.5

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2017.4

Language：English  

Venue：Yokohama   Country：Japan  

Cell image analysis is important for research and discovery in biology and medicine. In this paper, we present our cell tracking methods, which is capable of obtaining fine-grain cell behavior metrics. In order to address difficulties under dense culture conditions, where cell detection cannot be done reliably since cell often touch with blurry intercellular boundaries, we proposed two methods which are global data association and jointly solving cell detection and association. We also show the effectiveness of the proposed methods by applying the method to the biological researches.

Event date： 2017.4 - 2018.4

Language：English   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2016.9

Language：Japanese  

Country：Japan  

Event date： 2016.6 - 2016.7

Language：English  

Venue：Las Vegas   Country：United States  

A newly developed imaging technique called light-sheet laser microscopy imaging can visualize the detailed 3D structures of capillaries. Capillaries form complicated network structures in the obtained data, and this makes it difficult to model vessel structures by existing methods that implicitly assume simple tree structures for blood vessels. To cope with such dense capillaries with network structures, we propose to track the flow of blood vessels along a base-axis using a multiple-object tracking framework. We first track multiple blood vessels in cross-sectional images along a single axis to make the trajectories of blood vessels, and then connect these blood vessels to reveal their entire structures. This framework is efficient to track densely distributed vessels since it uses only a single cross-sectional plane. The network structure is then generated in the post-processing by connecting blood vessels on the basis of orientations of the trajectories. The results of experiments using a challenging real data-set demonstrate the efficacy of the proposed method, which are capable of modeling dense capillaries.

Event date： 2013.2

Language：English  

Venue：Innsbruck   Country：Austria  

Automated tracking of cell population is an important element of research and discovery in the biology field. In this paper, we propose a method that tracks cells under highly confluent conditions by using the candidate cell region detection-based association approach. Unlike conventional segmentation-based association tracking methods, the proposed method uses the tracking results from the previous frame to segment the cell regions at the current frame. First, candidate cell regions are detected, and while there may be many false positives, there are very few false negatives. Next, optimized detection results are selected from the candidate regions and associated with the tracking results of the previous frame by resolving a linear programming problem. We quantitatively evaluated the proposed method using a variety of sequences. Results showed that our method has a better tracking performance than conventional segmentation-based association methods.

Event date： 2011.8 - 2011.9

Language：English  

Venue：Boston, MA   Country：United States  

The wound healing assay in vitro is widely used for research and discovery in biology and medicine. This assay allows for observing the healing process in vitro in which the cells on the edges of the artificial wound migrate toward the wound area. The influence of different culture conditions can be measured by observing the change in the size of the wound area. For further investigation, more detailed measurements of the cell behaviors are required. In this paper, we present an application of automatic cell tracking in phase-contrast microscopy images to wound healing assay. The cell behaviors under three different culture conditions have been analyzed. Our cell tracking system can track individual cells during the healing process and provide detailed spatio-temporal measurements of cell behaviors. The application demonstrates the effectiveness of automatic cell tracking for quantitative and detailed analysis of the cell behaviors in wound healing assay in vitro.

Event date： 2011.3 - 2011.4

Language：English  

Venue：Chicago, IL   Country：United States  

Automated visual-tracking systems of stem cell populations in vitro allow for high-throughput analysis of time-lapse phase-contrast microscopy. In these systems, detection of mitosis, or cell division, is critical to tracking performance as mitosis causes branching of the trajectory of a mother cell into the two trajectories of its daughter cells. Recently, one mitosis detection algorithm showed its success in detecting the time and location that two daughter cells first clearly appear as a result of mitosis. This detection result can therefore helps trajectories to correctly bifurcate and the relations between mother and daughter cells to be revealed. In this paper, we demonstrate that the functionality of this recent mitosis detection algorithm significantly improves state-of-the-art cell tracking systems through extensive experiments on 48 C2C12 myoblastic stem cell populations under four different conditions.

Event date： 2011.3 - 2011.4

Language：English  

Venue：Chicago, IL   Country：United States  

Automated cell tracking in populations is important for research and discovery in biology and medicine. In this paper, we propose a cell tracking method based on global spatio-temporal data association which considers hypotheses of initialization, termination, translation, division and false positive in an integrated formulation. Firstly, reliable tracklets (i.e., short trajectories) are generated by linking detection responses based on frame-by-frame association. Next, these tracklets are globally associated over time to obtain final cell trajectories and lineage trees. During global association, tracklets form tree structures where a mother cell divides into two daughter cells. We formulate the global association for tree structures as a maximum-a-posteriori (MAP) problem and solve it by linear programming. This approach is quantitatively evaluated on sequences with thousands of cells captured over several days.

Event date： 2011.1

Language：English  

Venue：Kona, HI   Country：United States  

We present several algorithms for cell image analysis including microscopy image restoration, cell event detection and cell tracking in a large population. The algorithms are integrated into an automated system capable of quantifying cell proliferation metrics in vitro in real-time. This offers unique opportunities for biological applications such as efficient cell behavior discovery in response to different cell culturing conditions and adaptive experiment control. We quantitatively evaluated our system's performance on 16 microscopy image sequences with satisfactory accuracy for biologists' need. We have also developed a public website compatible to the system's local user interface, thereby allowing biologists to conveniently check their experiment progress online. The website will serve as a community resource that allows other research groups to upload their cell images for analysis and comparison.

Event date： 2010.4

Language：English  

Venue：Rotterdam   Country：Netherlands  

Cell segmentation in microscopy imagery is essential for many bioimage applications such as cell tracking. To segment cells from the background accurately, we present a pixel classification approach that is independent of cell type or imaging modality. We train a set of Bayesian classifiers from clustered local training image patches. Each Bayesian classifier is an expert to make decision in its specific domain. The decision from the mixture of experts determines how likely a new pixel is a cell pixel. We demonstrate the effectiveness of this approach on four cell types with diverse morphologies under different microscopy imaging modalities.

Event date： 2010.4

Language：English  

Venue：Rotterdam   Country：Netherlands  

We present a method for robustly detecting hematopoietic stem cells (HSCs) in phase contrast microscopy images. HSCs appear to be easy to detect since they typically appear as round objects. However, when HSCs are touching and overlapping, showing the variations in shape and appearance, standard pattern detection methods, such as Hough transform and correlation, do not perform well. The proposed method exploits the output pattern of a ring filter bank applied to the input image, which consists of a series of matched filters with multiple-radius ring-shaped templates. By modeling the profile of each filter response as a quadratic surface, we explore the variations of peak curvatures and peak values of the filter responses when the ring radius varies. The method is validated on thousands of phase contrast microscopy images with different acquisition settings, achieving 96.5% precision and 94.4% recall.

Event date： 2002.7

Language：English  

Venue：Frankfurt   Country：Germany  

This paper presents a design technique which is used to realize associative memories via cellular neural networks. The proposed method can store every prototype vector as a memory vector and maximize the areas of basin of attraction of memory vectors in a certain sense. The network parameters are obtained by solving optimization problems known as generalized eigenvalue problems. Simulation results prove that our method is better than the existing ones.

Language：English  

Venue：Tokyo   Country：Japan  

Language：English  

Venue：Tokyo   Country：Japan  

Language：English  

Country：Other  

Photoacoustic (PA) imaging has been gaining attention as a new imaging modality that can non-invasively visualize blood vessels inside biological tissues. In the process of imaging large body parts through multi-scan fusion,alignment turns out to be an important issue,since body motion degrades image quality. In this paper,we carefully examine the characteristics of PA images and propose a novel registration method that achieves better alignment while effectively decomposing the shot volumes into low-rank foreground (blood vessels),dense background (noise),and sparse complement (corruption) components on the basis of the PA characteristics. The results of experiments using a challenging real data-set demonstrate the efficacy of the proposed method,which significantly improved image quality,and had the best alignment accuracy among the state-of-the-art methods tested.

Language：Japanese  

Language：Japanese  

Language：Japanese  

Language：English  

Language：Japanese  

Language：Japanese  

Language：Japanese  

Language：Japanese  

Language：Japanese  

Language：Japanese  

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (scientific journal)  

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (scientific journal)  

Type：Competition, symposium, etc. 

Type：Competition, symposium, etc. 

Type：Competition, symposium, etc. 

Type：Competition, symposium, etc. 

Type：Competition, symposium, etc. 

Type：Competition, symposium, etc. 

Type：Peer review 

Number of peer-reviewed articles in foreign language journals：3

Proceedings of International Conference Number of peer-reviewed papers：18

Type：Competition, symposium, etc. 

Type：Competition, symposium, etc. 

Type：Competition, symposium, etc. 

Type：Competition, symposium, etc. 

Type：Peer review 

Number of peer-reviewed articles in foreign language journals：4

Proceedings of International Conference Number of peer-reviewed papers：20

Type：Competition, symposium, etc. 

Type：Competition, symposium, etc. 

Number of participants：7,500

Type：Competition, symposium, etc. 

Type：Peer review 

Number of peer-reviewed articles in foreign language journals：2

Proceedings of International Conference Number of peer-reviewed papers：14

Type：Competition, symposium, etc. 

Type：Peer review 

Number of peer-reviewed articles in foreign language journals：3

Number of peer-reviewed articles in Japanese journals：3

Proceedings of International Conference Number of peer-reviewed papers：16

Proceedings of domestic conference Number of peer-reviewed papers：2

Type：Competition, symposium, etc. 

Type：Competition, symposium, etc. 

Type：Peer review 

Number of peer-reviewed articles in foreign language journals：2

Number of peer-reviewed articles in Japanese journals：1

Proceedings of International Conference Number of peer-reviewed papers：9

Proceedings of domestic conference Number of peer-reviewed papers：1

Type：Peer review 

Number of peer-reviewed articles in foreign language journals：1

Number of peer-reviewed articles in Japanese journals：3

Proceedings of International Conference Number of peer-reviewed papers：13

Proceedings of domestic conference Number of peer-reviewed papers：0