Artificial Intelligence Meets Indoor Low-Cost Sensing for Healthcare and Monitoring
| Project/Area Number |
23K16950
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 61030:Intelligent informatics-related
|
|---|
| Research Institution | Keio University |
|---|
Principal Investigator |
ブーアジジ モンデル 慶應義塾大学, 理工学部(矢上), 特任講師 (10867675)
|
|---|
| Project Period (FY) |
2023-04-01 – 2026-03-31
|
| Project Status |
Granted (Fiscal Year 2023)
|
| Budget Amount *help |
¥3,770,000 (Direct Cost: ¥2,900,000、Indirect Cost: ¥870,000)
Fiscal Year 2025: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2024: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2023: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
|---|
| Keywords | Sensing for Healthcare / Artificial Intelligence / Non-contact sensors |
|---|
| Outline of Research at the Start |
FY2023: Person identification and Activity detection using 2D Lidar and stereoscopic IR imaging.
FY2024: Multiple Person identification Activity detection using 3D Lidar
FY2025: Vital signal monitoring using fusion between 3D Lidars and MIMO Doppler Sensor
|
| Outline of Annual Research Achievements |
So far, we have conducted several experiments on heart rate estimation and Human activity recognition in indoor environment using wireless sensors.
Mainly, Doppler Sensors, MIMO FMCW sensors and 2D/3D Lidars were used in the experiments.
The data collected during the experiments were then processed through signal processing techniques and Machine learning. The developed algorithm allowed us to obtain accurate results in both heart rate estimation and activity recognition. The obtained results were promising, and thus, were published in a few journal papers, presented in international and domestic conferences and discussed in workshops.
We have published over 5 journal papers and 5 international conference papers. However, newer results were obtained and are yet to be published as well.
|
| Current Status of Research Progress |
Current Status of Research Progress
1: Research has progressed more than it was originally planned.
Reason
Thanks to the availability of data and the data acquisition that went smoothly, the research has been advancing quite smoothly. So far, several research results have been published in International conferences and Journals.
The data acquired will be investigated further, and newer data will be collected as well.
The computers purchased allowed to train deep neural network models in short time, which allowed to obtain good results quickly.
|
| Strategy for Future Research Activity |
We plan to continue collecting data for our next experiments. To do so, we will acquire an advance Eye tracker and conduct experiments to identify fatigue, health problems such as depression as well as dementia using this device.
Nonetheless, we are exploring the potentials of using a MIMO FMCW radar for activity recognition, Heart rate and Blood Pressure estimation.
|
Report
(1 results)
Research Products
(10 results)
