Kyushu University Academic Staff Educational and Research Activities Database
List of Presentations
Toru Chikui Last modified date:2019.12.19

Associate Professor / Division of Maxillofacial Diagnostic and Surgical Sciences / Department of Dental Science / Faculty of Dental Science


Presentations
1. #Panyarak Wannakamon, Chikui Toru, Tokumori Kenji, Yamashita Yasuo, Yoshiura Kazunori, Utility of diffusion kurtosis model in differential diagnosis of orofacial tumors, 第47回日本磁気共鳴医学会大会, 2019.09, Main text :
Objectives: We aimed to present the utility of DK in orofacial region by comparing the parameters of DK and ME by means of the goodness of fit, the correlations of the parameters of two methods, and the effectiveness in differential diagnosis among various kinds of orofacial tumors.

Methods: One hundred and thirty-one patients underwent MR examination including EPI-DWI with 4 b-values (0, 500, 1000 and 1500 s/mm2). The ADC was obtained by the ME model. The DK parameters were estimated as apparent diffusion coefficient of Gaussian distribution (Dapp) and apparent kurtosis coefficient (Kapp). The Akaike Information Criterion was used to evaluate the goodness of fit. Then, the correlations of two methods were evaluated. These parameters were compared among cyst, benign and malignant tumors.

Results: The DK showed significantly better goodness of fit than the ME model (p<0.001). Dapp and ADC showed strong positive correlation (ρ=0.922) while Dapp and Kapp presented a strong negative correlation (ρ = -0.764). So, ADC and Kapp presented in a similar pattern (ρ=-0.92). Regarding the differential diagnosis, all parameters indicated significant differences among three groups (p<0.05). Dapp and ADC of cysts were the highest, while those of malignant tumors were the lowest. Conversely, Kapp showed opposing results as 0.450.60, 1.510.40 and 0.940.24 of cyst, benign and malignant tumors, respectively.
Conclusion: Given the better goodness of fit and the characteristic parameters of each category, the DK model is useful for the differential diagnosis in the orofacial tumors.
.
2. Wannakamon Panyrak, Toru Chikui, Kenji Tokumori, Yasuo Yamashita, Kazunori Yoshiura, Comparison among mono exponential model, IVIM model and gamma distribution model for the orofacial tumors: preliminary study, 日本歯科放射線学会第60回学術大会, 2019.06, Objectives: The first purpose is to compare the goodness of fit among mono-exponential (ME) model, IVIM model and Gamma distribution (GD) model and evaluate the correlation among these parameters. The second was to compare them between benign and malignant tumors.

Methods: 47 patients underwent 3T-MRI examination including TSE-DWI sequences with 6 b-values (0-1000 s/mm2). ADC was obtained by ME model, D, D*, and f were obtained by IVIM model. The shape parameter k, scale parameter θ and fractions of diffusion (ƒ1, ƒ2, ƒ3 which represented cell component, parenchyma and perfusion component, respectively) were estimated by the GD model. The Akaike Information Criterion was used for the evaluation of the goodness of fit. Then, the correlation among these parameters and the differences of the parameters between benign and malignant tumors were evaluated.

Results:
1. Both GD and IVIM models showed better goodness of fit than ME model.
2. ƒ1 showed the negative correlations with ADC (𝞀=-0.85) and D (𝞀=-076).
3. ƒ3 showed the positive correlation with ADC (𝞀=0.46), D (𝞀=0.30) and f (𝞀=0.65).
4. The malignant tumor showed significantly lower D, lower k and higher ƒ1 value, which suggested the higher cellularity.

Conclusions:
Given the better goodness of fit and the correlation of parameters between GD model and IVIM mode, GD model might be helpful for the interpretation of the DWIs.
.
3. #Wannakamon Panyarak, Toru Chikui, @Yasuo Yamashita, Kazunori Yoshiura, Image quality and ADC assessment in turbo-spin echo and echo-planar diffusion-weighted MR imaging of oral and maxillofacial tumors
, ESHNR2018, 2018.09, Purposes
The first purpose was to compare the distortion ratio (DR), the signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR) and ADC value between TSE-DWI and EPI-DWI. The second purpose was to demonstrate the location gap between TSE-DWI and T2WI and to estimate ADC in various kinds of disease.

Materials & Methods
Firstly, 42 patients underwent MRI using TSE-DWI and EPI-DWI. The DR, SNR, CNR and ADC were compared between the two techniques.
Secondly, 152 patients underwent MRI using TSE-DWI. The location gap along X- (direction of frequency encoding) and Y-axes (direction of phase encoding) was defined as the distance from the centroids of the lesion on DWIb1000 to that of T2WI. We also estimated ADC values of the lesions.

Results
38 out of 42 cases were analyzed. The DR of TSE-DWI was significantly less than that of EPI-DWI. The SNR as well as CNR of TSE-DWI were significantly higher than that of EPI-DWI. The ADC values of TSE-DWI were significantly higher than those of EPI-DWI in palatine tonsils and submandibular glands. However, ADC values of both sequences showed no significant differences in lesions and parotid glands.
In the group of TSE-DWI, 127 out of 152 cases were analyzed. The deviations along X-axis were 0.53±1.34 in lesion and 0.25±0.56 mm in spinal cord and that along the Y-axis were 0.34±1.61 and 0.73±0.82 mm, respectively. Regarding the ADC values, cysts and cystic-like lesions presented the highest values followed by benign tumors and squamous cell carcinoma, while malignant lymphoma showed the lowest value.

Conclusions
TSE-DWI can produce less distortion and higher image quality than EPI-DWI, while TSE-DWI possibly exhibits different ADC values. Therefore, the TSE-DWI could be good alternative to EPI-DWI and is useful for the differential diagnosis.
.
4. Wannakamon Panyarak, Toru Chikui, Yasuo Yamashita, Kazunori Yoshiura, Image quality and ADC assessment in turbo-spin echo and echo-planar diffusion-weighted MR imaging of oral and maxillofacial tumors, 日本歯科放射線学会第59回学術大会, 2018.05, Purposes
The first was to compare the distortion ratio (DR) and the signal-to-noise ratio (SNR) between TSE-DWI and EPI-DWI. The second was to demonstrate the image distortion between TSE-DWI and T2 WI and to estimate ADC in various kinds of diseases.

Materials & Methods
At first, 42 patients underwent MRI using TSE-DWI and EPI-DWI. The DR, SNR, and ADC were compared between the two techniques.
Secondly, 160 underwent MRI using TSE-DWI. The distortion along X- and Y-axes was defined the distance from the centroids of the lesion on DWIb1000 to that of T2WI. We also estimated ADC values of the lesions.

Results
DR of TSE-DWI was significantly less than that of EPI-DWI and SNR of TSE-DWI was significantly higher than that of EPI-DW. ADC values of palatine tonsils and submandibular glands in TSE-DWI were significantly higher than those in EPI-DW.
The deviations along X-axis were 0.57±1.30 mm and that along the Y-axis were 0.27±1.74 mm, therefore, the image distortion was very small. Regarding the ADC values, cysts presented the highest values followed by vascular lesion and benign tumors, while malignant lymphoma showed the lowest values.

Conclusions
TSE DWI can produce less distortion and higher SNR than TSE-DWI, while TSE-DWI possibly exhibits different ADC values. Therefore, the TSE-DWI could be good alternative to EPI-DWI and is useful for the differential diagnosis..
5. Wannakamon Panyarak, Toru Chikui, Yasuo Yamashita, Kazunori Yoshiura, Image quality assessment in turbo-spin echo diffusion-weighted MR imaging and the correlation between the ADC value and the histological features of head and neck tumors, ASHNR 2017, 2017.09, Purpose
Many studies have supported the utility of diffusion-weighted images (DWIs) in the differential diagnosis. However, Echo-planar imaging (EPI) DWIs are susceptible to inhomogeneity of the magnetic field, which limits their usefulness in the head and neck region. Thus, the first purpose of this study was to evaluate the image distortion of Turbo-spin echo (TSE)-DWIs in comparison to anatomical images. The second purpose was to demonstrate the apparent diffusion coefficient (ADC) values in various types of head and neck lesions.

Materials & Methods
In all cases, MRI was performed using a clinical 3T whole-body MRI system (Philips Healthcare, Best, Netherlands). A total of 99 cases were enrolled. In each case, MRI was performed and T2-weighted images (T2WIs) and TSE-DWIs (b-value, 0 and 1000 s/mm2 [DWIb0 and DWIb1000]) were obtained.
At first, the DWI was reoriented onto a T2WI to match the matrix size and slice thickness. Then, we carefully drew the lesion, the bilateral Masseter muscles (MMs) and the spinal cord and defined them as regions of interest (ROIs) on a DWIb0, a DWIb1000 and a T2WI. The centroids of the ROIs were obtained. The differences along the X-axis (the direction of frequency encoding) and the Y-axis (the direction of phase encoding) were measured among these centroids. For the second purpose, an ADC map was obtained using DWIb0 and DWIb1000 images. All of the procedures were performed using the OsiriX Lite software program (Pixmeo SARL, Bernex, Switzerland) and the Image J software program (NIH, Maryland, USA).

Results
The 88 cases (89%) (benign tumor, n=20; malignant tumor, n=52; inflammation, n=9; cyst, n=7) could be evaluated by TSE-DWI .Eleven cases were excluded because their lesions were too small (n=8), or due to the presence of a severe metal artifact (n=3).
Focusing on the comparison of DWIb1000 images to T2WIs, the deviations along X-axis were 0.87±0.62 mm, 0.74±0.61 mm, and 0.42±0.38 mm (lesion, MMs, and spinal cord, respectively). Meanwhile, the deviations along the Y-axis were 1.37±1.28 mm, 1.44±0.99 mm, and 0.67±0.57 mm (lesion, MMs, and spinal cord, respectively). The deviation along the Y-axis was significantly larger than that along the X-axis in MMs and the spinal cord (p-value < 0.05); however, the absolute amount was not so large. In addition, the deviation that was observed between DWIb0 images and T2WIs was almost identical to that observed between DWIb1000 and T2WI (mean difference, <0.1 mm). Thus, the image distortion in the direction was not so severe.
Regarding the ADC values, cysts presented the highest values (1.91±0.59 x10-3 mm2/s) followed by benign tumors (1.46±0.44 x10-3 mm2/s), while malignant lymphoma showed the lowest values (0.86±0.12 x10-3 mm2/s). Thus, the ADC values obtained from TSE-DWIs were useful for evaluating these lesions.

Conclusions
TSE-DWIs were found to be immune to image distortion; thus, these images could be easily superimposed on anatomical images. Moreover, the ADC values that were obtained showed high diagnostic power in assessment of head and neck lesions.
.
6. Toru Chikui, Wannakamon Panyarak, Yasuo Yamashita, Toshiyuki Kawazu, Kazunori Yoshiura, The detection of recurrence of oral cancer using turbo spin echo-diffusion-weighted imaging , ASHNR 2017, 2017.09, Objective
Many studies have supported the utility of diffusion-weighted image (DWI) for detecting recurrence. However, Echo-planar imaging DWIs are susceptible to inhomogeneity of the magnetic field, which limits its usefulness in the head and neck region. The first objective was to evaluate the utility of turbo spin echo (TSE)-DWI for detecting recurrence of oral cancer. The second objective was to compare the observer performance between TSE-DWI and gadolinium-enhanced T1-weighted imaging (Gd-T1WI).
Materials and Methods
Forty-two patients underwent surgery for oral squamous cell carcinoma and postoperative MRI, including axial T2-weighted imaging (T2WI), TSE-DWI and Gd-T1WI. All MRI procedures were performed using a 3T MRI system (Philips Healthcare, Best, the Netherlands). Twenty-two patients were classified as the recurrence group, and 20 were classified as the non-recurrence group. First, the DWI images were automatically adjusted for the translational and rotational motion to achieve superimposition with the Gd-T1WI images. We then assumed the high-signal-intensity area on DWIb1000 images or the enhanced area on Gd-T1WI images to indicate signs of recurrence and calculated the sensitivity and specificity. Once the enhanced area was identified, we outlined this area as a region of interest, copied it onto the adjusted ADC map, and measured the ADC value. These procedures were performed using the PACS system (Volume Analyzer Synapse Vincent; FUJIFILM, Tokyo, Japan). Next, two observers (observer 1: 22 years experience, observer 2: 3 years experience) interpreted the presence of recurrence using a five-point scale. The first session included the T1WI, T2WI and Gd-T1WI images, and the second included the axial T1WI, T2WI, DWIb1000, ADC map and combined T2WI and DWIb1000 fusion images. We estimated the area under the curve (AUC) using a receiver operating characteristic analysis. The image interpretation was performed on a personal computer using the OsiriX Lite software (Pixmeo SARL, Bernex, Switzerland).
Results
The sensitivity was 81.8% on both DWI b1000 and Gd-T1WI, and the specificity was 75% on DWI b1000 and 55% on Gd-T1WI. Although the specificity of DWI b1000 was quite high, the T2 shine-through might have reduced it to some degree. Of the 27 cases with enhanced areas on Gd-T1WI, the ADC of non-recurrence (1.82±0.51×10−3 mm2/s) was significantly higher than that of recurrence (1.14±0.113×10−3 mm2/s) (P<0.0001, Wilcoxon’s test). The AUCs in the first session including Gd-T1WI were 0.74±0.08 and 0.66±0.08 for observers 1 and 2, respectively. The AUCs in the second session including DWI were 0.79±0.07 and 0.82±0.06 for observers 1 and 2, respectively. There was a significant difference in the findings of observer 2 between the two methods (P=0.02). Therefore, DWI yielded a higher AUC, even if they did not used the enhanced MR images. In addition, a low ADC value was found to be characteristic of recurrence, and a good fusion image on DWI along with anatomical images were believed to help observers interpret the imaging findings.
Conclusions
TSE-DWI was found to be immune to image distortion and proved useful for differentiating between recurrence and a postoperative state. Adding the TSE-DWI sequence improved the observer performance regarding the interpretation of recurrence.
.
7. Toru Chikui, Wannakamon Panyarak, Yasuo Yamashita, Shintaro Kawano, 吉浦 一紀, The detection of a recurrence of oral cancer using TSE-DWI, The 21th International Congress of Dental and Maxilla-Facial Radiology, 2017.04, Objective: To evaluate the feasibility of utilizing Turbo Spin Echo (TSE)-diffusion weighted imaging (DWI) to detect the recurrence of oral cancer.

Materials and Methods: Thirty-six patients who underwent surgery for oral cancer were enrolled. Recurrence was found in eighteen patients. Examinations were performed on a 3T-MR system and the acquired sequences consisted of T2W-TSE, TSE-DW (b values=0,1000s/mm2), and Gd-enhanced T1W-TSE. DW images were coregistered with Gd-enhanced T1W-TSE images. We evaluated the presence of a high signal area on DWI with a b value of 1000s/mm2 and that of the enhanced area on Gd-enhanced T1WI. If any enhanced areas were found, we drew an outline to identify the region of interest and then measured the ADC on the coregistered ADC map.

Results: The parallel displacement of the coregistration was -0.14±0.83, -0.12±0.80 and -0.34±0.56 mm along x, y and z axis, respectively. The rotation was 0.34±0.56°, - 0.13±0.58°, and , 0.29±0.40°in three axes. Therefore, the image misregistration between the two sequences is considered to be quite small. Three recurrences could not be detected on any sequence due to the small size of the lesion. The sensitivity of recurrence was 83% on both DWI and Gd-enhanced T1WI. Moreover, the high-signal areas on DWI coincided with the enhanced areas. The specificity was 83% on DWI and 72% on Gd-enhanced T1WI. The ADC of non-recurrence (1.90±0.14x10−3mm2/s) was significantly higher than that of recurrence (1.19±0.07x10−3mm2/s) (P=0.008).

Conclusions: TSE-DWI was thus found to be immune to image distortion and it was found to be useful for detecting recurrence. When contrast enhanced MRI can not be performed, then TSE-DW may be a powerful alternative diagnostic modality..
8. Tomoko Shiraishi, Toru Chikui, Kunihiro Miwa, Toyohiro Kagawa, Shoko Yoshida, Kenji Yuasa, Quantitative shear wave elastography for the differentiation between metastatic and inflammatory cervical lymph nodes., 11th Asian congress of oral and maxillo-facial Radiology, 2016.11.
9. Yoshitaka Kise, Toru Chikui, Yasuo Yamashita, Kouji Kobayashi, 吉浦 一紀, Eiichirou Ariji, Quantitative shear wave elastography for the differentiation between metastatic and inflammatory cervical lymph nodes., 11th Asian congress of oral and maxillo-facial Radiology, 2016.11.
10. Chikui T, Yamashita Y, Shimizu M, Yoshimura K, Estimation of the proton density fat fraction of the salivary gland, ASHNR 50th Annual Meeting, 2016.09, Purpose
The first purpose was to assess the value of a six-echo variant of the modified Dixon Quant (mDixon) for salivary gland fat fraction (FF) quantification against multi-echo T2 corrected MR spectroscopy which was used as the reference standard. The second goal was to apply mDixon to estimate the FFs and to compare them among the control group, Sjogren's syndrome (SS) group and IgG4 related disease (IgG4-RD) group.
Materials and Methods
To conduct this study, 10 healthy volunteers were enrolled and 20 parotid glands (PGs) were evaluated. Quantitative MRI was performed using the mDixon sequence in addition to single-voxel MRS to compare the mDixon-FF with MRS-FF. On the FF image obtained by the mDixon method, we measured the FF of the region of interest (ROI). For the MRS, five unsuppressed 1H spectra were acquired by the Stimulated Echo Acquisition Mode with different echo delays (TE=13-53ms). Time-domain quantification was performed using the AMARES algorithm from the MRUI software package. Relaxation times T2 of water and the four main different triacylglycerol components (5.3, 2.1,1.3,and 0.9 ppm) were determined. For the second purpose, we applied the mDixon sequence to 64 cases. Eleven patients were thus assigned to the SS group, 4 patients to the IgG4-RD group, and the remaining 49 patients to the control group with any abnormality of the salivary gland. The outlines of both the PGs and submandibular glands (SMGs) were traced to set the ROIs to estimate the FF. In the control group, we evaluated any correlations among the FF, height, weight, body mass index (BMI), cholesterol level, and serum triglyceride level. We also compared the FF among the three groups.
Results
The FF(mDixon) correlated closely with the FF(MRS) according to the following equation FF(mDixon)=1.06xFF(MRS)-2.6 with R2 of 0.99. Therefore, the FF(mDixon) was considered to be potentially appropriate for clinical use. The PG-FF of the control group positively correlated with the weight and BMI (?=0.32, P=0.024 and ?=0.59, P < 0.001, respectively). The SMG-FF also correlated with the weight, BMI and serum triglyceride levels (?=0.59,P=0.0228, ?=0.38,P=0.0067 and ?=0.52, 0.0004, respectively). The SS group had a higher PG-FF(50.3±1.7%) than both the control group (34.9±11.1%) and IgG4-RD group (35.2±21.7%) and a significant difference was found between the SS group and control group (P=0.0113, Steel-Dwass test). This trend was remarkable in the submandibular gland. The SS group had a significantly higher SMG-FF(41.1±25.9%) than both the control group (7.4±4.5%) and IgG4-RD group (3.9±3.2%) (P < 0.0001 and P=0.0258, respectively, Steel-Dwass test). Therefore, the SMG-FF was lower than the PG-FF in the control group (P < 0.001, paired t test), however fat may accumulate rapidly in SMG patients due to the progression of SS.
Conclusions
Estimating the FF of the salivary gland using mDixon was thus found to be highly accurate in comparison to the MRS method. Therefore, this modality is considered to be potentially appropriate for clinical application. The FF obtained by mDixon quantitatively revealed the characteristic fat accumulation (especially SMGs) in the SS group..
11. Estimation of the fat fraction of the salivary gland using mDixon Quant.
12. Yoshitaka Kise, Toru Chikui, Yashio Yamashita, Koji Kobayashi, Atsushi Takemura, Kazunori Yoshiura, Eiichiro Ariji, Salivary gland fat fraction estimated with a new MRI method, IADR, Objectives: Measurement of fat fraction (FF) in the salivary glands is important in such diseases as Sjogren'ssyndrome, and it can be performed with MR spectroscopy. However, there is a limit in clinical application due to the complicated manipulations. A new magnetic resonance imaging (MRI) method, named mDIXON Quant, is convenientfor clinical application. The aim of this study was to estimate FF in the salivary glands using two different methods,mDIXON Quant and MR spectroscopy, and then to investigate the clinical usefulness of mDIXON Quant incomparison with MR spectroscopy.Methods: Sixteen healthy volunteers were enrolled in this study. They received two types of MRI examination,mDIXON Quant (6-point DIXON method) and MR spectroscopy (PRESS Single TE method), to measure FF in theparotid and submandibular glands. Ten subjects were also scanned by PRESS Multi TE method to determine theeffect of T2 value. On mDIXON Quant images, the regions of interest were set on both sides of the parotid andsubmandibular glands using the obtained FF map to determine FF.Results: 1. High correlation was observed in FF measured by mDIXON Quant and PRESS (Single TE) (slope=1.07, R2=0.960). 2. T2 value of the fat peak calculated by PRESS (Multi TE) was greater than T2 value of the water peak. 3. Even in taking the effect of T2 value by Multi TE method into account, a high correlation between mDIXON Quantand PRESS (Multi TE) was observed (slope=0.594, R2=0.964).Conclusions: The mDIXON Quant method appeared to be effective for FF measurement in the salivary glandsbecause the values obtained by mDIXON Quant correlated well with those by MR spectroscopy..
13. Kazutoshi Okamura, Yutaka Yoshida, Chikui Toru, Kazunori Yoshiura, Evaluation of improvement effect on the iterative reconstruction method in hard tissue, The 10th Asian Congress of Oral and Maxillo Facial Radiology, 2014.11, Institution and Country
Department of Oral & Maxillofacial Radiology, Faculty of Dental Science, Kyushu University, Japan.
*Department of Health Sciences, Faculty of Medical Sciences, Kyushu University, Japan.

Correspondence e-mail
okamura@rad.dent.kyushu-u.ac.jp

Introduction
Iterative Reconstruction (IR) algorithms for Computed Tomography (CT) is widely used in clinical practice in late years. Although improvement in image quality and dose reduction have been reported, many were evaluated based on soft tissue.

Objectives
To evaluate improvement effect on the IR method in hard tissue like teeth and bone, with high-contrast object.

Materials and methods
Images of a high-contrast and a water cylinder phantom were acquired with Alexion four-row CT (Toshiba Medical Systems, Tokyo, Japan) at 120 kV tube voltage and five tube current levels. They were reconstructed with filtered back projection (FBP) and adaptive iterative dose reduction three-dimensional (AIDR 3D; Toshiba Medical Systems) reconstruction algorithms using the bone kernel. The slice thickness was 1.0mm and the reconstruction FOV was 70.2mm. The minimum separable hole size was estimated on the images of the high contrast phantom and the noise power spectrum (NPS) was measured on the images of the water cylinder phantom for FBP and for four levels of AIDR 3D.


Results
The NPSs of AIDR 3D were lower than that of FBP. The noise reduction effect was observed visually at low tube current. The minimum separable hole size was 0.5mm at 150 and 100mA, it raised by lowering tube current, and it was 1.0mm at 10mA. There ware no effect of AIDR 3D in detection size.

Conclusions and discussion
On these conditions, the effect of AIDR 3D did not lead to improvement in detection ability. At the time of evaluation of the hard tissue with low tube current, use of AIDR 3D may have an opposite effect. Evaluation with complicated structures like trabecular bone is required.
.
14. Tomoko Shiraishi, Chikui Toru, Marie Hashimoto, Kenji Yuasa, Evaluation of Malignant lymphoma(ML) using MRI and US, The 10th Asian Congress of Oral and Maxillo Facial Radiology, 2014.11, Introduction
Malignant lymphoma(ML)almost occurs in lymph node, but rare extranodal. It was frequently mistaken for another malignant tumor.
Objectives
To elucidate the characteristics of ML imaging features in head and neck region.
Materials and methods
We described 8 patients (5 male and 3 female, average age, 67 years, age range 20). 8 lesions were histopathologically diagnosed as ML (tongue, 1 buccal mucosa, 3 maxilla and palate, 3 submandibular, 1) between 2006 and 2012. MRI was performed all lesions. We evaluated the signal intensity (SI) of T1WI, T2WI and estimated the diffusion coefficient (ADC) with b-factors of 0,500,1000 sec/mm2. In four cased, the dynamic MRI study was performed and the time intensity curve (TIC) was obtained. US was performed 2 lesions. US end point was internal echogenicity, internal vascularity, posterior echo enhancement.
Results
All ML cases had low intensity on T1WI. On T2WI, SI ranged from middle to high intensity. Mean ADC values were 0.771±0.142 sec/mm2 for ML, which were significantly lower than squamous cell carcinoma (? ± ??sec/mm2 in our preliminary study). TIC type showed a rapid increase downward titration type or a rapid increase storm decrease type.
Conclusions and discussion 
On both T1WI and T2WI, SI was not specific for ML. However, ADC was extremely low and the most characteristic finding, which suggested the small amount of extracellular space.
.
15. Chikui Toru, Erina Kitamoto, Masahiro Ohga, Kazunori Yoshiura, The pharmacokinetic analysis of the dynamic contrast –enhanced MRI in maxillofacial region., The 10th Asian Congress of Oral and Maxillo Facial Radiology, 2014.11, Introduction
Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is widely performed, but cannot elucidate the underlying physiological conditions. A pharmacokinetic analysis of the DCE-MRI findings may provide such information.
Objectives:
To elucidate the characteristics of tumors using a pharmacokinetic analysis.
Materials and Methods
A total of 135 tumors were examined[91 squamous cell carcinomas (SCC), 24 malignant lymphomas (ML), 10 malignant salivary gland tumors (MSGTs) and 10 pleomorphic adenomas (Pleo)]. We applied the Tofts and Kermode model and obtained three parameters: the influx forward volume transfer constant into the extravascular extracellular space (EES) from the plasma (Ktrans), the fractional volume of EES (ve) and the fractional volume of plasma (vp). We compared the parameters among the four types of tumors and also evaluated the correlation between the TMN stage and the parameters in SCC.
Results
ML had a significantly lower ve than the other categories (SCC; P<0.0001, MSGT; P=0.049 and Pleo; P=0.0001) and it also had the lowest Ktrans. Pleo had the highest ve value, significantly higher than that of SCC (P=0.0033), and had the lowest vp, which was significantly different from SCC (P=0.018). A small amount of EES and low permeability were characteristic of ML, while a large amount of EES and poor vascularity were characteristic of Pleo. In SCC, the Ktrans decreased with the progression of the N stage, and the Ktrans of N2 cases was lower to that of No cases (P = 0.0028).
Conclusions
The pharmacokinetic analysis clarified the physiological conditions of the tumors.
.
16. Attempt for the estimation of the concentration of the Gd-contrast medium. Phantom study and clinical application..
17. Quantitative analysis of sonographic images of the salivary gland: Comparison between the sonographic findings and the sialographic findings.
18. Digital intraoral radiography system for filmless environment.
19. Evaluation on changes of T2 values and ADCs of the salivary glands by acid stimulation..
20. Attempt for the estimation of the concentration of the Gd-contrast medium. Phantom study and clinical application..
21. T1 measurement by the Look-Locker sequence. The phantom study and the application to the head and neck lesion.
22. Clinical practice guidelines for oral and maxillofacial radiology.
23. Sonographic texture characterizationof the parotid gland of the patient with the xerostomia.
24. The role of the Ultrasonography for the diagnoses of the patients with oral cancer..
25. Sonographic texture characterization of salivary gland tumors by fractal analyses.
26. Sonographic texture characterization of salivary gland tumors by fractal analyses.
27. Change of microangiostructure and hemodynamics in lymph node metastases in rabbits, animal model for Doppler sonography.
28. The diagnosis of masses in the submandibular region.
29. Intraoral sonography for the tongue cancer after interstitial brachytherapy.
30. The state of imaging diagnosis of cervical lymph nodes..
31. Intraoral sonography of the mobile tongue:.
32. The early detection of cervical lymph node metastasis: The analysis of the changes of the findings of power doppler sonographic images..
33. Change of sonographic findings on cervical lymph nodes before and after preoperative radiotherapy.
34. Change of microangiostructure and hemodynamics in lymph node metastases in rabbits, animal model for Doppler sonography.