| Unique ID issued by UMIN | UMIN000061814 |
|---|---|
| Receipt number | R000070735 |
| Scientific Title | Development and Validation of a Machine Learning-Based Skeletal Muscle Mass Estimation Model Using Smartphone-Captured Lower-Leg Images: Comparison with Calf Circumference |
| Date of disclosure of the study information | 2026/06/05 |
| Last modified on | 2026/06/05 23:51:19 |
Development and Validation of a Skeletal Muscle Mass Estimation Model Using Smartphone-Captured Lower-Leg Images
Smartphone-based Muscle Index from Lower-leg Study
Development and Validation of a Machine Learning-Based Skeletal Muscle Mass Estimation Model Using Smartphone-Captured Lower-Leg Images: Comparison with Calf Circumference
Lower-Leg Image SMI Study
| Japan |
Healthy adults
| Rehabilitation medicine | Adult |
Others
NO
The purpose of this study is to validate a deep learning-based method for estimating skeletal muscle index (SMI) using lateral and posterior lower-leg images captured with smartphones and digital cameras. In addition, differences in predictive performance according to imaging device and imaging direction will be investigated, and comparison with calf circumference, a conventional surrogate measure of skeletal muscle mass, will be performed to evaluate the usefulness of image-based assessment.
Efficacy
Confirmatory
Explanatory
Not applicable
Concordance between skeletal muscle mass estimated from lower-leg images and skeletal muscle mass measured by bioelectrical impedance analysis (InBody 470), assessed using Lin's concordance correlation coefficient (CCC)
1. Prediction error of the skeletal muscle mass estimation model using smartphone-captured lower-leg images, assessed by mean absolute percentage error (MAPE) and root mean squared error (RMSE)
2. Agreement between skeletal muscle mass estimated from smartphone-captured lower-leg images and skeletal muscle mass measured by bioelectrical impedance analysis, assessed using Bland-Altman analysis (mean difference and 95% limits of agreement)
3. Comparison of predictive performance (CCC, MAPE, and RMSE) between the smartphone image-based skeletal muscle mass estimation model and the calf circumference-based estimation model
4. Comparison of skeletal muscle mass estimation performance according to imaging device (smartphone or digital camera) and imaging direction (lateral or posterior lower-leg view), assessed using CCC, MAPE, and RMSE
Interventional
Single arm
Non-randomized
Open -no one is blinded
Uncontrolled
NO
NO
Institution is not considered as adjustment factor.
NO
No need to know
1
Prevention
| Device,equipment |
Lateral and posterior lower-leg images will be captured once using a smartphone and a digital camera. On the same day, skeletal muscle mass will be measured once using bioelectrical impedance analysis (InBody 470), and calf circumference will be measured once.
| 18 | years-old | <= |
| Not applicable |
Male and Female
1. Individuals who provide written informed consent to participate in the study
2. Healthy adults aged 18 years or older
3. Individuals capable of maintaining standing and sitting positions during assessment
4. Individuals able to undergo lower-leg image acquisition and bioelectrical impedance analysis
1. Individuals who have difficulty maintaining a standing position
2. Individuals with edema or other conditions affecting the lower leg
3. Individuals with implanted medical devices such as cardiac pacemakers
4. Individuals with any limb amputation
5. Individuals with metallic implants or fixation devices in the body
6. Individuals who are pregnant or may be pregnant
7. Individuals with a history of fractures or ligament injuries of the lower limbs
8. Individuals with a history of central or peripheral nervous system disorders
9. Individuals deemed unsuitable for participation in the study by the principal investigator or co-investigators
100
| 1st name | Hiroo |
| Middle name | |
| Last name | Matsuse |
Kurume University Hospital
Department of Rehabilitation
830-0011
67 Asahi-machi, Kurume-shi, Fukuoka 830-0011, Japan
0942-35-3311
matsuse_hiroh@kurume-u.ac.jp
| 1st name | Hiroo |
| Middle name | |
| Last name | Matsuse |
Kurume University Hospital
Department of Rehabilitation
830-0011
67 Asahi-machi, Kurume-shi, Fukuoka 830-0011, Japan
0942-35-3311
matsuse_hiroh@kurume-u.ac.jp
Kurume University
None
Other
Clinical Research Center, Kurume University Hospital
67 Asahi-machi, Kurume-shi, Fukuoka 830-0011, Japan
0942-65-3749
i_rinri@kurume-u.ac.jp
NO
| 2026 | Year | 06 | Month | 05 | Day |
Unpublished
100
No longer recruiting
| 2025 | Year | 01 | Month | 30 | Day |
| 2025 | Year | 01 | Month | 30 | Day |
| 2025 | Year | 02 | Month | 14 | Day |
| 2025 | Year | 08 | Month | 31 | Day |
| 2025 | Year | 08 | Month | 31 | Day |
| 2026 | Year | 06 | Month | 05 | Day |
| 2026 | Year | 06 | Month | 05 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/icdr_e/ctr_view.cgi?recptno=R000070735