This page organizes papers related to artificial intelligence published around the world. This page is summarized using Google Gemini and is operated on a non-profit basis. The copyright of the paper belongs to the author and the relevant institution. When sharing, simply cite the source.
This paper presents a multi-modal smart home platform designed for continuous self-rehabilitation of stroke patients. This platform integrates wearable sensing, environmental monitoring, and adaptive automation. It classifies motor recovery stages with up to 94% accuracy using plantar pressure insoles and a machine learning pipeline, and enables hands-free appliance control using an eye-tracking module, camera, and microphone. A hierarchical IoT architecture fuses data, and the embedded large-scale language model (LLM) agent, Auto-Care, provides real-time intervention. In a study of stroke patients, the average user satisfaction rating improved from 3.9 in a traditional home environment to 8.4 when the system was fully operational.
Takeaways, Limitations
•
Takeaways:
◦
Presenting an integrated smart home platform for self-rehabilitation of stroke patients.
◦
Effective combination of wearable sensors, environmental sensors, and LLM-based automation.
◦
Provides quantitative tracking of exercise recovery stages and hands-free appliance control.
◦
Improving user satisfaction and offering the possibility of personalized care through real-time intervention.
◦
Potential expansion into broader neurorehabilitation and geriatric care fields.
•
Limitations:
◦
Lack of further information on specific technical details and algorithm performance.
◦
Results of a study with a limited number of 20 subjects.
◦
Long-term use and efficacy in diverse patient populations are needed.
◦
Additional considerations regarding data privacy and security are needed.
◦
Further research is needed on the system's cost, complexity, and user acceptability.
