Abstract:
People suffering from chronic diseases require continuous support in monitoring their nutrition, diagnostic tests, medication, and daily activity tracking. Given the low ratio of patients to healthcare providers, it becomes infeasible to provide one-to-one support to patients. Furthermore the existing online medical consultation platforms are costly for regular approach. Addressing these issues, we propose to monitor and assist patients suffering from chronic disease using a novel Al-based and IoT-supported digital twin platform. A digital twin of a patient grows with the patient, and it helps in continuous and remote patient monitoring. Further, the digital twin enables the creation of patient-specific personalized treatment models, enabling doctors to conduct virtual simulations of the suitability of certain drugs and procedures. The data collected from the digital twin is fed to machine learning models for intelligent analysis, feedback, and support. The proposed solution incorporates five essential machine learning models using novel algorithms for drug recommendation, chronic disease stage detection, nutrition tracking and recommendation, patient activity tracking, and patient data anonymization. Addressing patients' lack of motivation to participate in emerging patient monitoring frameworks, we incorporate an incentive mechanism rooted in Non-Fungible Tokens (NFTs) to encourage active participation in patients, which also has the added benefit of helping patients to store their historical medical data securely.