Adaptive Treatment Assisting System for Patients Using Machine Learning

• Published in: 2019 Sixth International Conference on Social Networks Analysis, Management and Security (SNAMS) pp. 460 - 465
• Main Authors: Naeem, Muddasar, Coronato, Antonio, Paragliola, Giovanni
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.10.2019
• Subjects: disabilities; Drugs; Machine learning; Mathematical model; medicine; Reinforcement learning; Security; Social networking (online); Visualization
• DOI: 10.1109/SNAMS.2019.8931857

Machine learning techniques have achieved a lot of success in many healthcare related aspects including personalized treatment, medical imaging, diagnostic systems etc. In present work, we propose a medicine reminder system which can assist patients in their treatment process at home. In case of treatment at home of patients with different physical and/or mental disabilities, a key challenge is the choice of the proper message for each specific patient taking into account his/her needs. To this aim, the proposed system can use a variety of messages (i.e audio, visual and textual) to communicate with patients taking into account their specific physical and/or mental disabilities. We apply Upper Confidence Bound method which enables the guidance system to adapt according to a patient`s mental and physical skills. To demonstrate the feasibility of the proposed treatment assisting system, a sample treatment plan are implemented in an simulated experiment. The results show that our proposed system through interactions with the patient of simulated physical and mental skills is capable of selecting right mode of message which will lead the patient to the right medicine box.