| Unique ID issued by UMIN | UMIN000063027 |
|---|---|
| Receipt number | R000072149 |
| Scientific Title | Human Movement as a Digital Vital Sign: A Standardized Framework for Video-Based Artificial Intelligence Motion Analysis |
| Date of disclosure of the study information | 2026/09/23 |
| Last modified on | 2026/09/23 12:26:43 |
Video-based AI motion analysis of physical function in patients with cardiovascular disease
AI-MOVE study
Human Movement as a Digital Vital Sign: A Standardized Framework for Video-Based Artificial Intelligence Motion Analysis
AI-MOVE (AI MOtion analysis as a Vital sign Evaluation) study
| Japan |
Cardiovascular disease requiring hospitalization, including heart failure, ischemic heart disease, valvular heart disease, arrhythmia, pulmonary hypertension, and peripheral artery disease
| Cardiology | Rehabilitation medicine |
Others
NO
To evaluate the feasibility of video-based artificial intelligence motion analysis during Short Physical Performance Battery (SPPB) testing performed as part of routine inpatient care in patients hospitalized for cardiovascular disease, and to explore associations between video-derived motion features and conventional physical performance measures, frailty, activities of daily living, musculoskeletal conditions, and post-discharge outcomes, in order to establish human movement as a quantitative digital vital sign.
Others
Feasibility of video-based AI motion analysis and exploration of its associations with physical function and clinical outcomes
Exploratory
Explanatory
Not applicable
Proportion of enrolled patients with analyzable video data at the first SPPB assessment during hospitalization, calculated for each SPPB component (standing balance, 4-m gait, and repeated chair stands). A video is defined as analyzable when the required key points are detected throughout the task and the whole body remains in view. The prespecified target is at least 80% for each component.
1. Completion rate of video-recorded SPPB tasks and reasons for unanalyzable videos (at the first assessment during hospitalization)
2. Distribution of video-derived motion features (at the first assessment during hospitalization)
3. Associations between video-derived motion features and SPPB score, gait speed, frailty, activities of daily living, musculoskeletal conditions, and clinical characteristics
4. Comparison of information obtained from frontal and lateral views, to define the minimum requirements for future single-view acquisition
5. In a subset of at least 30 patients, agreement between video-derived gait speed and chair-stand time and stopwatch-measured values (intraclass correlation coefficient, Bland-Altman analysis), and reproducibility on repeated analysis of the same videos by independent analysts
6. Exploratory associations between video-derived motion features and outcomes at 6 and 12 months and up to 5 years after discharge (decline in activities of daily living, new or worsened care-needs certification, falls, institutionalization, rehospitalization, cardiovascular events, all-cause death)
Observational
| 20 | years-old | <= |
| Not applicable |
Male and Female
1. Aged 20 years or older
2. Hospitalized at Kumamoto University Hospital for cardiovascular disease (heart failure, ischemic heart disease, valvular heart disease, arrhythmia, pulmonary hypertension, peripheral artery disease, etc.) or undergoing inpatient cardiac rehabilitation
3. Scheduled to undergo the Short Physical Performance Battery (SPPB) or part of it during hospitalization or before discharge
4. Able to perform at least one video-recordable physical performance task
5. Able to provide written informed consent
1. Aged under 20 years
2. Active malignancy, severe infection, or severe trauma
3. Physical performance testing considered unsafe because of hemodynamic instability, severe respiratory compromise, acute neurological deterioration, or other unstable medical conditions
4. Unable to perform any of the 3 SPPB components because of severe physical disability
5. Severe cognitive impairment or communication difficulty precluding understanding of the study procedures
6. Participation judged inappropriate by the attending physician or the investigator
150
| 1st name | Yasushi |
| Middle name | |
| Last name | Matsuzawa |
Kumamoto University Hospital
Department of Cardiovascular Medicine
860-8556
1-1-1 Honjo, Chuo-ku, Kumamoto 860-8556, Japan
096-373-5175
matsuzawa-y@kumamoto-u.ac.jp
| 1st name | Yasushi |
| Middle name | |
| Last name | Matsuzawa |
Kumamoto University Hospital
Department of Cardiovascular Medicine
860-8556
1-1-1 Honjo, Chuo-ku, Kumamoto 860-8556, Japan
096-373-5175
matsuzawa-y@kumamoto-u.ac.jp
Kumamoto University
Yasushi Matsuzawa
none
Self funding
Kumamoto University IRB
1-1-1 Honjo, Chuo-ku, Kumamoto 860-8556, Japan
096-373-5657
iyks-iji@jimu.kumamoto-u.ac.jp
NO
熊本大学病院(熊本県)
| 2026 | Year | 09 | Month | 23 | Day |
Unpublished
Enrolling by invitation
| 2025 | Year | 09 | Month | 12 | Day |
| 2025 | Year | 09 | Month | 12 | Day |
| 2026 | Year | 01 | Month | 01 | Day |
| 2035 | Year | 03 | Month | 31 | Day |
| 2035 | Year | 03 | Month | 31 | Day |
| 2035 | Year | 03 | Month | 31 | Day |
| 2035 | Year | 03 | Month | 31 | Day |
This registration covers the artificial intelligence motion analysis component of a prospective cohort study of patients hospitalized for cardiovascular disease (Kumamoto University, Approval No. 3338). The primary endpoint of the overall cohort is a composite of cardiovascular events or all-cause death during follow-up of up to 5 years.
| 2026 | Year | 09 | Month | 23 | Day |
| 2026 | Year | 09 | Month | 23 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000072149