Systematic analysis demonstrates 94% classification accuracy for fall detection in older adults, highlighting the viability of real-time deep learning for home monitoring.
Gait monitoring applications are deployed in the areas of sports, fall detection, health diagnostics, and treatment. Elders who live alone or with limited manual assistance need constant monitoring. The elderly suffer from falls due to gait abnormality which may result from aging, neurological disorders, and certain side effects of medications and many such factors. In health diagnostics, the researchers have proposed various solutions for analyzing human gait posture recognition and detection by analyzing the data collected from wearable sensors or applying deep learning algorithms on the video inputs. This paper presents a systematic approach of various methods employed by researchers and practitioners to observe gait patterns by employing deep learning algorithms to the image frame obtained from a real-time environment for fall detection in the elderly community. Results show that LSTM approach provides 94 % classification accuracy with minimal false alarms.
No takes yet. Share an insight, caveat, or question.
Elavarasi et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: