Related Experiment Video
Updated: Aug 6, 2026
13:35
Segmentation and Linear Measurement for Body Composition Analysis using Slice-O-Matic and Horos
Published on: March 21, 2021
Multi-component exercise intervention methods for intelligent assisted diagnosis of sarcopenia in the elderly based
Xiangkun Yang1, Soh Kim Geok2, Chin Yit Siew3
1Department of Sports Studies, Faculty of Educational Studies, University Putra Malaysia, Serdang, 43400, Selangor, Malaysia.
Scientific Reports
|July 16, 2026
Summary
This study developed an AI framework using CNN-LSTM models for sarcopenia diagnosis and personalized exercise interventions. The integrated approach achieved high accuracy, improving muscle mass and strength in older adults.
Area of Science:
- Geriatric Medicine
- Artificial Intelligence in Healthcare
- Biomedical Engineering
Background:
- Sarcopenia poses a significant public health challenge due to population aging.
- Current diagnostic methods are inefficient and lack early risk prediction capabilities.
- Standardized interventions fail to address individual muscle function variations in older adults.
Purpose of the Study:
- To develop an integrated framework for intelligent sarcopenia diagnosis and personalized intervention.
- To address the disconnect between sarcopenia diagnosis and intervention strategies.
- To leverage multimodal data and deep learning for improved geriatric care.
Main Methods:
- Integrated multimodal data: muscle ultrasound, body composition, muscle strength, and clinical records.
- Developed an improved Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) model.
- Applied Z-score normalization for data standardization and employed CNN for spatial and LSTM for temporal feature extraction.
Main Results:
- The CNN-LSTM model achieved 92.3% diagnostic accuracy and 0.95 AUC for sarcopenia.
- Early identification of mild sarcopenia reached an 89.5% recall rate.
- Personalized intervention led to significant improvements in muscle mass (12.7%), strength (18.2% grip, 15.5% knee), and walking speed (0.22 m/s).
Conclusions:
- The proposed framework offers a low-cost, practical solution for large-scale sarcopenia screening and precision health management.
- This deep learning approach advances precision-oriented geriatric healthcare for age-related diseases.
- The closed-loop diagnosis-intervention system demonstrated superior outcomes compared to standardized programs with minimal adverse events.
