Kyushu University Academic Staff Educational and Research Activities Database
Researcher information (To researchers) Need Help? How to update
Yoshinori Katayama Last modified date:2018.06.12



Graduate School
Undergraduate School


E-Mail
Homepage
http://bie.inf.kyushu-u.ac.jp/~yosinori/
Homepage of Yoshinori Katayama .
Phone
092-802-3768
Fax
092-802-3767
Academic Degree
Doctor of Engineering
Field of Specialization
Pattern Recognition
Outline Activities
After the study of sentence level speech recognition under the constraint of context free grammar,
now the study about on-line character recognition and brain infomatics.
Research
Research Interests
  • Signal processing of EEG aiming for BCI
    keyword : Brain-Computer Interface, BCI, EEG signal processing, EEG, Artifact, TMS, Transcranial Magnetic Stimulation, electroencephalogram
    2008.05Theme: Signal processing against electroencephalogram for brain computer interface Keywords: brain computer interface, BCI, electroencephalogram, EEG, EEG signal processing, artifact, transcranial magnetic stimulation,TMS Outline: Signal processing and pattern recognition against electroencephalogram for constructing the brain-computer interface (BCI). Signal processing around electroencephalogram such as modeling of the specific artifact of applying the transcranial magnetic stimulation (TMS). .
  • Learning method of the Cube search HMM and modeling of the stroke HMM in the online character recognition
    keyword : online character recognition, stroke order free, cube search, Hidden Markov Model (HMM), hook
    2006.04Theme: Learning method of the Cube search HMM and modeling of the stroke HMM in the online character recognition. Keyword: online character recognition, stroke order free, cube search, Hidden Markov Model (HMM), hook Outline: Improve the recognition performance in the popular writing of the online character recognition with basically stroke-order free by including the statistical stroke order information within the training data, which is achieved by applying the HMM to the cube search of fully stroke-order free system, that is called cube HMM. Stroke HMM is the stroke level modeling and the base system against the cube HMM. Investigating the model design of the stroke HMM which is more adequate and robust against the specific noise such as hook elements. .
Academic Activities
Papers
1. Kazuki Onikura, Yoshinori Katayama, Keiji Iramina, Evaluation of a Method of Removing Head Movement Artifact from EEG by Independent Component Analysis and Filtering, Advanced Biomedical Engineering, DOI:10.14326 / abe.4.67, 4, 67-72, 2015.03, Artifacts that contaminate electroencephalography (EEG) signals make it difcult to analyze EEG. The aim of this study was to removal artifacts on EEG, especially those caused by motion, to measure EEG in unconstrained situations. In a previous study, head movements were detected by an accelerometer, and motion artifact components were separated from the recorded EEG by independent component analysis (ICA). This method is effective for reducing the effect of artifacts, but has a risk that EEG components are also removed.
In this paper, we introduce an improved artifact removal method based on ICA and filtering. EEG were decomposed by ICA, and a Pearson’s correlation coefficient was calculated between each independent component and each hybrid accelerometer value to distinguish artifact components. Artifact components were then high-pass filtered. In this study, subjects were instructed to move their heads randomly, while keeping their eyes closed. The previous method was adapted using 1, 2, 3, 5 and 10 s to find the most suitable epoch to minimize the mean absolute amplitude of the cleaned EEG. Then, using this epoch, the proposed method was compared with the previous method by frequency analysis.
Low frequency power (0.1–3 Hz) was normalized to unity because most power caused by motion artifacts exists in the low power band. If the normalized theta (4–8Hz), alpha (8–13Hz) and beta (13–40Hz) powers of cleaned EEG are higher than that of raw EEG, this indicates that the effect of motion artifacts is small and EEG components are retained.
The results obtained from theta and alpha power comparison showed that the proposed method performed better than the previous method. This result suggests that the proposed artifact removal method is more effective to reduce the effect of artifacts while retaining the EEG components..
2. Latest Report about Biomagnetism, including Transcranial Magnetic Stimulation.
3. Fumiyoshi Matsusaki, Yoshinori Katayama, Keiji Iramina, Influence of TMS Coil Orientation in the Simulation of Neuronal Excitation by TMS Using an Axon Model and Cerebral Cortex Model, Advanced Biomedical Engineering, DN/JST.JSTAGE/abe/1.55, 1, 1, 55-59, 2013.06, Transcranial magnetic stimulation (TMS) allows non-invasive and painless stimulation of local
cerebral nerves using eddy current generated by electromagnetic induction with a TMS coil. Although TMS is
used in various fields, which area of the brain is stimulated is not known because of the complicated structure of
the organ. In this study, we simulated neuronal excitement by TMS using the finite element method. First, we
designed a brain sulcus model consisting of cerebrospinal fluid, gray matter and white matter, using 0.5 mm cube
elements. To improve calculation accuracy, cube element size was set to 0.5/3 mm only in regions near the
boundary surface. Second, we applied TMS stimulation to the model in different conditions. We used coil radii of
10, 20, 30 and 40 mm, and coil orientation at 0°, 30°, 45°, 60° and 90°, which is defined as the angle between the
orientation of the electric field and the axon. Finally, we calculated the membrane potential and compared the
results obtained under different conditions. We found that membrane potential changed rapidly at the white
matter and gray matter interface when the coil radius was over 20 mm and coil orientation was within 60°
between the orientation of the electric field and the axon. These results provide useful information on appropriate
TMS parameters for effective stimulation of target area in the brain..
4. Y. Katayama K. Iramina, Fitting and Eliminating to the TMS Induced Artifact on the Measured EEG
by the Equivalent Circuit Simulation Improved Performance
, 5th Kuala Lumpur International Conference on Biomedical Engineering 2011: Biomed 2011, 20-23 June 2011, Kuala Lumpur, Malaysia (Ifmbe Proceedings), 10.1007/ 978-3-642-21729-6, 35, 519-522, 2011.06, Transcranial magnetic stimulation (TMS) is the non-invasive stimulus method to the brain by inducing the eddy current within the brain from the TMS coil placed outside the scalp. The induced eddy current stimulates the nerve circuit causes to suppress the brain activity partially in time and space, and it is applied to detect the brain functions etc. When TMS is applied to the brain while measuring the electroencephalography (EEG), the induced artifact caused by TMS covers on the EEG, it is called the TMS artifact. The amplitude of the TMS artifact is generally too large to omit diagnosing the EEG. Therefore some methods have been proposed to negate the TMS artifact from the EEG including the TMS artifact using the EEG only. We proposed a new method for describing the shape of the induced artifact in the EEG applied TMS using two equivalent circuit models, the TMS equipment model and the equivalent circuit model of bioelectric measurement system, under some simplified approximations.
This paper shows that some attempts are applied to the proposed method to improve the performance of fitting and eliminating the TMS artifact. One is to derive the strict solution of the TMS artifact by omitting some simplified approximations. Other is the countermeasure against errors calculating inverse matrix while estimat-ing parameters of fitting shape of the TMS artifact. One attempt is found that the strict solution improves the shape fitting but not so far from the approximated solu-tion. Other attempt improves stability of the simulation. These attempts are found that some time parameters which determine the time-transition such as “Tc” should separately treat against other time constant parameters which are derived from circuit parameters. Moreover, considering the residual component this is not described on the circuit model.
.
5. Yoshinori Katayama, Keiji Iramina, Equivalent Circuit Simulation of the Induced Artifacts Resulted from Transcranial Magnetic Stimulation on Human Electroencephalography, IEEE Trans. on Magnetics, Vol. 45, Issue 10, pp.4833 - 4836, 2009.10.
6. Takahiro Matsunaga, Tetsuya Fukuta, Yoshinori Katayama, Keiji Iramina, Analysis of evoked response and induced response in alpha wave and gamma wave during visual attention, Brain Topography and Multimodal Imaging, Proceedings of the 18th International Congress on Brain Electromagnetic Topography , 23-26, 2009.09.
7. Y.Katayama, S.Uchida, H.Sakoe , Stochastic Model of Stroke Order Variation, ICDAR 2009, pp.803-807, 2009.07.
8. Y.Katayama, S.Uchida, H.Sakoe, A New HMM for On-Line Character Recognition Using Pen-Direction and Pen-Coordinate Features, The 19th ICPR2008, CDROM, 2008.12.
Presentations
1. Yuki Noguchi, Miki Kaneko, Keita Higashi, Yasushi Miyagi, R. Murayama, Keiji Iramina, Yoshinori Katayama, Evaluation for Treatment of Deep Brain Stimulation by Pronation and Supination of Forearms using Wireless Sensors
, 7th Biomedical Engineering International Conference (BMEiCON 2014), 2014.11.
2. Extended Equivalent Circuit Model to Eliminate the Artifact Induced by Transcranial Magnetic Stimulation on the Measured EEG
.
3. Approximated Circuit Model of the TMS Artifact and Its Elimination.
4. TMS artifact model and EEG signal processing.
5. Performance Improvement about the Simulation of the TMS Artifact on the EEG .
6. Simulation of the induced TMS artifact in the measured Electroencefalogram(EEG) .
Membership in Academic Society
  • The Institute of Electrical Engineers of Japan
  • The Institute of Electrical and Electronics Engineers (IEEE)
Awards
  • The Best Paper Award of IEICE 2008 (at 23 May 2009)
    Title: An HMM Representing Stroke Order Variations and Its Application to Online Character Recognition