Akademik Veri Yönetim Sistemi
IEEE Transactions on Computational Social Systems, 2026 (SCI-Expanded, Scopus)
With the rapidly growing global population, the early detection of diseases in older individuals has become a critical challenge, as they often face difficulties accessing physical clinics. Conventional healthcare solutions struggle to effectively address the prodromal phase of chronic conditions, hindering timely identification and mitigation. This research proposes a state-of-the-art framework integrating artificial intelligence (AI), multimodal large language models (MLLMs), and metaverse technologies to significantly improve early diagnosis and personalized healthcare for elderly patients, by leveraging MLLMs to analyze and synthesize a wide range of data, including physiological signals, behavioral patterns, and medical imaging. Our proposed framework provides a comprehensive and precise understanding of an elderly patient's health status. We introduce an innovative algorithm that streamlines data from these diverse modalities and applies fine-tuning on an open-source AI model to predict disease progression. The metaverse component of our framework furnishes a compelling and accessible environment, using an AI-driven avatar (meta-doctor) for real-time consultations, diagnoses, and personalized health recommendations. The research results highlight the potential of combining MLLMs and metaverse technologies to boost diagnostic accuracy, improve remote patient engagement, and enhance healthcare outcomes for senior citizens.