Revolutionizing Elderly Home Care with AI Systems: Towards Smart Monitoring and Personalized Assistance in Rehabilitation Therapy

As the age of the global population increases, the demand for innovative and practical solutions to support elderly care has attracted significant interest. Recent advances in digital technologies and computational science offer promising opportunities to extend the continuum of care at patients’ homes. This dissertation investigates the potential role of artificial intelligence (AI) methods and technology can have in rehabilitation medicine. The hypothesis is that AI can provide solutions for smart monitoring and assistance in home-based rehabilitation programs, empowering patients to actively participate in individualized treatment, improve self-management, and autonomously adhere to personalized long-term therapies. State-of-the-art deep learning algorithms for computer vision and pattern recognition were analyzed and proposed in the monitoring and evaluation of human motion behavior during physical therapy exercises. The research also explored bias mitigation and resilience techniques in algorithm implementation to support the integration of effective and equitable AI technologies into future clinical applications. At the same time, the ethical challenges posed by these technological advancements were examined, focusing on the core principles and strategies to ensure the fairness of AI technologies that support clinical decision-making. A critical evaluation was also conducted on the regulatory framework within which these technologies could be effectively implemented and governed. This thesis pioneers the development and application of AI-driven technologies in rehabilitation medicine. Its aim is to create more accessible, responsive, and impactful healthcare solutions to address the pressing global need for high-quality, cost-effective, and efficient treatments to manage the growing healthcare demands of non-communicable chronic diseases driven by population aging.