2023 Fiscal Year Research-status Report
Integrated deep learning model for personalized transcranial magnetic stimulation
| Project/Area Number |
22K12765
|
|---|
| Research Institution | University of Hyogo |
|---|
Principal Investigator |
Rashed Essam 兵庫県立大学, 情報科学研究科, 教授 (60837590)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
平田 晃正 名古屋工業大学, 工学(系)研究科(研究院), 教授 (00335374)
ゴメスタメス ホセデビツト 千葉大学, フロンティア医工学センター, 准教授 (60772902)
|
|---|
| Project Period (FY) |
2022-04-01 – 2025-03-31
|
| Keywords | Deep learning / brain stimulation / TMS / Segmentation |
|---|
| Outline of Annual Research Achievements |
In this year, we have have conducted a comprehensive training process for the developed deep learning models. The training consider optimizing the process of TMS focalization of specific brain region (motor cortex) and how to estimate stimulation parameters in different scenarios. The SHARM dataset (https://arxiv.org/abs/2309.06677) developed last year is used in the training process with TMS data obtain from our research collaborators.
After the initial training, the model parameters are optimized using several validation studies to achieve superior network performance. We have evaluated different versions of the network architecture to validate potential variations such as adding attention layers and include BN/dropout layers. Now, we are in the phase of preparing publications.
|
| Current Status of Research Progress |
Current Status of Research Progress
1: Research has progressed more than it was originally planned.
Reason
The research is progress smoothly than planned. We initially plan to complete three work packages in the first year (development of the deep learning model, data collection, and TMS simulation). Which are all successfully completed. Furthermore, we have reported results through several presentations and invited talks. Work package 4 that include network training (scheduled for FY2023) was partially completed in FY2022. Moreover, we have completed Work Package 5 (validation and optimization) which was planned to be completed within the third year.
|
| Strategy for Future Research Activity |
The research plan for FY2024 include two work packages (Result reporting) and (deployment of open-source software). We have made some conference publications and already submitted one journal paper to international journal and currently under review. Further publications is expected based on results we already have in hands.
Additional experiment will be conducted for extension of the achieved results in terms of TMS focal point optimization. Also, we will prepare documentation for the open-source software to ease sharing with research community.
|
| Causes of Carryover |
Some amount of the budget was actually planned for equipment purchase but we have found that it can be shifted to next year to fit more with the project progress and activities.
|
