Markov Decision-based Recommender System for Sleep Apnea Patients

Few decades ago, wellness management systems were not in the position to give salutary to their users. One of the possible reasons is inefficient resources and minimum technological infrastructure which do not allow a comprehensive structure pertaining to specific user. Sleep apnea is one such problem which is related to the permanent condition that involve stagnation of breathing. Although incurable, it can be minimized by maintaining a healthy lifestyle. Motivated from this, in this paper, we propose a health manager directive system to investigate the precise medical condition of sleep apnea. A recommender system is used which suggests the healthy lifestyle schedule to reduce the apnea severity in a patient. A Probabilistic Markov model (PMM) is used to adhere the activities based on time consumption in different activities performed by the patient. We evaluated the recommendation cycle on three patients to demonstrate the reductions in apnea cycles by indicating sound sleep patterns. Numerical results show that the proposed recommendation System suggest a relative improvement in sleep quality for all patients as compared to pre-existing expensive detection and relief schemes for sleep apnea patients.