| Unique ID issued by UMIN | UMIN000059576 |
|---|---|
| Receipt number | R000068134 |
| Scientific Title | Development and validation of a restless behavior detection model using Inertial Measurement Unit sensors and healthy person data |
| Date of disclosure of the study information | 2025/12/01 |
| Last modified on | 2025/10/29 10:52:38 |
Development and validation of an artificial intelligence model to detect restless movements using motion sensors
AI motion sensor study for detecting restlessness
Development and validation of a restless behavior detection model using Inertial Measurement Unit sensors and healthy person data
Inertial Measurement Unit-based Restless behavior Estimation and Sensing Technology Study: IMU-REST STUDY
| Japan |
None
| Not applicable |
Others
NO
The objective is to develop (train) a convolutional neural network (CNN) model that detects agitated behavior from motion data recorded by an IMU sensor using healthy subject data, and to evaluate its initial feasibility for future clinical application in intensive care units.
Efficacy
F1 Score for Target Location Estimation Using Convolutional Neural Networks (CNN)
Sensitivity
Specificity
Accuracy
AUC
etc.
Observational
| 18 | years-old | <= |
| 65 | years-old | >= |
Male and Female
healthy person
Individuals under the age of 18
Individuals with underlying musculoskeletal conditions or a history of collagen diseases
Individuals with a pacemaker implant
Individuals who did not consent to the study
45
| 1st name | Ryuto |
| Middle name | |
| Last name | Yokoyama |
Yamagata University Faculty of Medicine
Department of Emergency and Citical Care Medicine
990-9585
2-2-2 Iida Nishi, Yamagata City, Yamagata Prefecture
0236331122
ryusi0311@gmail.com
| 1st name | RYUTO |
| Middle name | |
| Last name | YOKOYAMA |
Yamagata University Faculty of Medicine
Department of Emergency and Critical Care Medicine
9902321
2-2-2 Iida Nishi, Yamagata City, Yamagata Prefecture
08055544587
ryusi0311@gmail.com
Yamagata Univercity
RYUTO YOKOYAMA
Yamagata Univercity
Other
Ethics Review Committee, Faculty of Medicine, Yamagata University
2-2-2 Iida Nishi, Yamagata City, Yamagata Prefecture
0236331122
ryusi0311@gmail.com
NO
| 2025 | Year | 12 | Month | 01 | Day |
Unpublished
Preinitiation
| 2025 | Year | 10 | Month | 28 | Day |
| 2025 | Year | 12 | Month | 01 | Day |
| 2028 | Year | 01 | Month | 01 | Day |
Recruitment of Participants and Explanation/Consent
This study will involve healthy individuals who have received a thorough explanation of the study's purpose and content and have provided written consent.
Sensor Attachment
IMU sensors will be safely attached to the participant's limbs and waist.
Performance of Movement Tasks
Under the researcher's guidance, subjects will perform multiple physical movements (e.g., elbow flexion/extension, sitting, trunk flexion/extension) that we consider "restless movements" within safe limits. Research staff will be present and monitor during the movements to avoid risks such as falls.
Data Acquisition and Storage
Data obtained from the IMU sensors will be collected and securely stored.
Data Analysis and Machine Learning Model Development
The acquired data will be organized and preprocessed to extract features for each movement. This data will be used to train a CNN model and develop an automatic detection model for agitated behavior.
Model Validation
The performance of the trained model will be evaluated using validation and test data. Validity will be assessed based on metrics such as accuracy, recall, and F1 score.
| 2025 | Year | 10 | Month | 29 | Day |
| 2025 | Year | 10 | Month | 29 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/icdr_e/ctr_view.cgi?recptno=R000068134