โก Quick Summary
This study explored the feasibility of a smartphone-based gait assessment protocol to classify age and sex in low-resource settings. The results demonstrated a successful extraction of gait parameters, achieving a maximum mean area under the receiver operating characteristic curve of approximately 0.90 for sex classification and 0.70 for age classification.
๐ Key Details
- ๐ Participants: 155 individuals (Thailand: 59, India: 96)
- โ๏ธ Technology: MediaPipe algorithm for pose estimation
- ๐งฉ Features extracted: 109 parameters related to joint distances, angles, and walking speed
- ๐ Performance: Sex classification AUC ~0.90, Age classification AUC ~0.70
๐ Key Takeaways
- ๐ฑ Smartphone technology can be effectively used for gait assessment in low-resource settings.
- ๐ก Machine learning models can classify biological features such as age and sex based on gait parameters.
- ๐ฅ High feasibility was observed with pose parameters extracted from 93.5% of video recordings.
- ๐ Classification performance was influenced by the number of features, clothing, and pose estimation quality.
- ๐ Potential applications include disease screening and longitudinal health monitoring.
- ๐ Future research is needed to explore more complex health-related conditions using this protocol.
๐ Background
Gait assessment is a crucial tool for evaluating health risks, particularly in older adults. However, it remains underutilized in low-resource settings where access to advanced medical technologies is limited. This study aimed to bridge that gap by utilizing a simple walking protocol combined with smartphone video capture to extract health-related gait signals, focusing on the classification of age and sexโtwo fundamental biological factors that significantly influence health outcomes.
๐๏ธ Study
Conducted as a cross-sectional study, this research involved 155 participants from Thailand and India. Each participant performed a straightforward walking protocol while being recorded on smartphones. The MediaPipe algorithm was employed to extract 109 features related to gait, including joint distances, angles, and walking speed. The study aimed to assess the feasibility of using these features for machine learning classification of age and sex.
๐ Results
The study successfully extracted pose parameters from 145 out of 155 video recordings, representing a 93.5% success rate. Among these participants, 64.8% were female, and 37.9% were aged 65 years or older. The machine learning models demonstrated comparable performance, with sex classification achieving a maximum mean area under the receiver operating characteristic curve of approximately 0.90 and age classification reaching around 0.70.
๐ Impact and Implications
The findings of this study have significant implications for public health, particularly in low-resource settings. By utilizing a simple smartphone-based gait assessment protocol, healthcare providers can potentially screen for health risks associated with aging and biological sex. This approach not only enhances accessibility to health assessments but also opens avenues for further research into more complex health-related conditions, paving the way for improved health monitoring and disease prevention strategies.
๐ฎ Conclusion
This study highlights the promising potential of using machine learning and smartphone technology for gait assessment in low-resource settings. The ability to classify age and sex based on gait parameters could lead to significant advancements in health monitoring and disease screening. As we look to the future, further research is essential to explore the broader applications of this innovative protocol in healthcare.
๐ฌ Your comments
What are your thoughts on the use of smartphone technology for gait assessment? We would love to hear your insights! ๐ฌ Leave your comments below or connect with us on social media:
Video-Based Gait Assessment Using Machine Learning to Classify Age and Sex in Low-Resource Settings: Cross-Sectional Study.
Abstract
BACKGROUND: Gait assessment is an important tool for evaluating health risks in older adults but remains underused in low-resource settings. We explored the feasibility of using a low-cost, simple walking protocol with smartphone video capture to extract health-related gait signals by classifying sex and age. Sex and age are fundamental biological factors linked to most health- and aging-related outcomes. Establishing baseline classification performance provides justification for future exploration of more complex health-related conditions using this protocol.
OBJECTIVE: This study aimed to assess whether pose parameters derived from smartphone-based gait videos can be used by machine learning models to classify age and sex.
METHODS: A cross-sectional study was conducted with 155 participants (Thailand: n=59, 38.1%; India: n=96, 61.9%). Participants performed a simple walking protocol while being recorded using smartphones. Pose estimation was conducted using the MediaPipe algorithm to extract 109 features related to joint distances, angles, and walking speed. For feasibility assessment, we calculated the proportion of recordings for which pose estimation could be extracted. Elastic-net logistic regression and histogram-based gradient boosting classifiers were used for analysis. Model performance was evaluated using 5-fold cross-validation. Outcomes were sex (male vs female) and age group (aged<65 vs โฅ65 y).
RESULTS: Pose parameters were successfully extracted from 145 (93.5%) of the 155 video recordings. Among the 145 participants, 94 (64.8%) were female, and 55 (37.9%) were aged 65 years or older. The 2 analytic models demonstrated comparable performance. Sex classification achieved a maximum mean area under the receiver operating characteristic curve of approximately 0.90 (SD 0.06), whereas age classification achieved a maximum mean area under the receiver operating characteristic curve of approximately 0.70 (SD 0.09). Classification performance was primarily influenced by the number of features used, clothing characteristics, and the quality of pose estimation.
CONCLUSIONS: This simple smartphone-based gait assessment protocol was able to extract meaningful pose parameters and classify biological features (age and sex). Further studies are warranted to evaluate its potential utility for disease screening, risk stratification, and longitudinal health monitoring.
Author: [‘Aramrat C’, ‘Mallinson PAC’, ‘Inkaew P’, ‘Seepheung P’, ‘Wiwatkunupakarn N’, ‘Buawangpong N’, ‘Birk N’, ‘Lieber J’, ‘Bhogadi S’, ‘Mahajan H’, ‘Banjara SK’, ‘Kulkarni B’, ‘Kinra S’, ‘Angkurawaranon C’]
Journal: JMIR Form Res
Citation: Aramrat C, et al. Video-Based Gait Assessment Using Machine Learning to Classify Age and Sex in Low-Resource Settings: Cross-Sectional Study. Video-Based Gait Assessment Using Machine Learning to Classify Age and Sex in Low-Resource Settings: Cross-Sectional Study. 2026; 10:e76755. doi: 10.2196/76755