Related Experiment Video
Updated: May 2, 2026
09:41
Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
Published on: April 21, 2023
2.8K
Grip Strength Estimation Using Input Data From a Commodity Smartphone: Model Development and Validation Study
Komei Tajima1, Kaori Ikematsu2, Toshiya Isomoto2
1Department of Information and Computer Science, Faculty of Science and Technology, Keio University, Yokohama, Kanagawa, Yokohama, Japan.
JMIR Human Factors
|April 30, 2026
Summary
This study shows smartphones can estimate grip strength using touch and sensor data, offering a convenient alternative to dynamometers for monitoring muscle health.
Area of Science:
- Biomedical Engineering
- Human-Computer Interaction
- Health Monitoring
Background:
- Grip strength is a key indicator for muscle deterioration, sarcopenia, and neurological conditions.
- Traditional grip strength measurement requires specialized, inaccessible dynamometers.
Purpose of the Study:
- To develop and validate a smartphone-based method for estimating grip strength.
- To eliminate the need for dedicated grip strength measurement devices.
Main Methods:
- Collected grip strength and smartphone interaction data (tapping, flicking, dragging) from 21 adults.
- Developed a predictive regression model using touch and inertial sensor data.
- Evaluated model accuracy via random split, leave-one-user-out, and few-day calibration validation.
Main Results:
- Random split evaluation showed high accuracy (MAE 2.62 kg, MAPE 8.91%).
- Leave-one-user-out validation yielded a MAPE of 15.08%.
- Personalized calibration improved accuracy, reducing MAPE to 11.64% after 4 days; subjective workload was low.
Conclusions:
- Smartphones offer a viable and accessible tool for daily grip strength monitoring.
- The proposed method provides a convenient alternative to traditional dynamometers.
- This technology supports pervasive health monitoring without user burden.
