Utilizing Assimilated Whale Optimization and Deep Q-Learning to Interfere with Sensed Input Classification Model for Remote Patient Monitoring

Remote Patient Monitoring (RPM) systems, which are based on the Internet of Medical Things (IoMT), offer real-time data and insights on patients' illnesses without requiring frequent in-person visits to healthcare facilities. The wearable sensors are in charge of measuring people's psychological vitals at various intervals in order to determine their exact health state. In order to solve the continuous and discrete data extraction problems in RPM, this paper presents the Interfering Input Classification Model (IICM). The suggested approach combines deep-Q learning with whale optimization to identify connections and classify various interval data. The whales' search process ending is used to separate the felt data into discrete and continuous categories. The maximal discreteness of such classed data is verified, as is its relationship to earlier sequences. Using continuous clinical range correlations and Q-learning based on various state changes, this connectivity is carried out. The suggested model serves as an example of continuous and discrete signals, or data, for the purpose of anomaly identification and diagnosis suggestion. As a result, the suggested model may consistently increase accuracy and the classification ratio while introducing less errors.

Author(s): Sayyed Johar* and GR Manjula

Abstract | PDF

Share This Article