2020 Fiscal Year Annual Research Report
強縦断的な生体医学信号の深層学習と健康関連の人工知能応用
| Project/Area Number |
19F19081
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
山本 義春 東京大学, 大学院教育学研究科(教育学部), 教授 (60251427)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
QIAN KUN 東京大学, 教育学研究科(研究院), 外国人特別研究員
|
|---|
| Project Period (FY) |
2019-10-11 – 2022-03-31
|
| Keywords | Signal Processing / Internet of Things / Artificial Intelligence |
|---|
| Outline of Annual Research Achievements |
In summary, we have achieved plenty of milestones during the FY2020. We introduced a novel paradigm that utilises the usage recorded data from smart appliances to analyse the elderly’s behaviour in a long duration. This non-intrusive approach can facilitate the combination of artificial intelligence and internet of things (AIoT) for making a more convenient and flexible life for the ageing population. This work was published online by the top journal IEEE Internet of Things Journal (with an impact factor of 9.936). Moreover, we systematically summarised the scenarios, data modalities, and methodologies for AIoT-enabled applications for the specific elderly group. We also indicated the benchmarks and limitations of the existing studies and gave our perspectives on future work. This article has been accepted and will be published by the prestigious journal IEEE Signal Processing Magazine (with an impact factor of 11.350). A comprehensive review was done and invited to be published by the IEEE Journal of Biomedical and Health Informatics (with an impact factor of 5.223). This review article concluded the state-of-the-art of audio-based methods for localising the snore site in the past three decades. In addition, we formed a team to collaboratively propose a novel approach for monitoring the confirmed COVID-19 patients on their sleep quality, fatigue, and anxiety. The relevant studies were published in the IEEE Internet of Things Journal and ISCA INTERSPEECH conference.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
We are now working towards transferring our methods to more general purposes, e.g., monitoring the drowsiness of drivers via the spontaneous physical activity data. We are also investigating the advanced data augmentation methods for coping with the data scarcity challenge among the several applications, e.g., the audio-based COVID-19 diagnosis problem. Furthermore, we are exploring the optimal time-frequency methods for analysing the body sound signals. Some preliminary results have already been achieved in recent study on heart sound analysis work.
|
| Strategy for Future Research Activity |
We will continuously collect more human behaviour data in near future, which may include multiple modalities, e.g., audio, video, and wearable sensors. We also want to build an explainable AI system for understanding the human behaviour in a high-level paradigm, which can benefit improving the model’s generalisation for multiple tasks.
|
-
-
-
-
-
-
[Journal Article] Predicting Group Work Performance from Physical Handwriting Features in a Smart English Classroom.2021
Author(s)
Meishu Song, Kun Qian, Bin Chen, Keiju Okabayashi, Emilia Parada-Cabaleiro, Zijiang Yang, Shuo Liu, Kazumasa Togami, Ichiro Hidaka, Yueheng Wang, Bjorn W. Schuller, and Yoshiharu Yamamoto.
-
Journal Title
Proceedings of ICDSP
Volume: -
Pages: 1～5
Peer Reviewed / Int'l Joint Research
-
-
-
-
