An efficient cat hunting optimization-biased ReLU neural network for healthcare monitoring system

Nowadays, various social media platforms, as well as wearable sensor devices, play a significant role in collecting data from patients for effective healthcare monitoring. However, continuous monitoring of patients using wearable sensor devices generates a huge amount of data and can be a complicated task to analyze efficiently. Therefore, this paper proposes a novel framework called Biased ReLU Neural Network-based Cat Hunting Optimization (BRNN-CHO) to classify the health condition of the patient. The proposed system consists of five phases: the data source phase, data collection phase, data pre-analysis phase, data pre-processing phase, and data classification phase. In the data source layer, various heterogeneous data including various medical records, social media platforms, and wearable sensor devices are addressed. The second data collection phase collects data regarding patients with blood pressure and diabetes. Then, in the data storage phase, the data collected from various medical records, social media platforms, and wearable sensor devices are uploaded to the big data cloud center. After uploading, the data is pre-analyzed and pre-processed to extract unwanted data. Finally, the pre-processed data is classified using BRNN-CHO to determine the health condition of the patient related to mental health, diabetes, and blood sugar, as well as blood pressure, with an enhanced accuracy rate. Experimental analysis is carried out and compared with various state-of-the-art techniques to determine the efficiency of the proposed system. When compared in terms of accuracy, F-measure, precision, and recall, the proposed BRNN-CHO model offers higher performance. The accuracy, recall, precision, and F1-measure of the proposed model are nearly equal to 95%, 90%, 92%, and 93%. The Root mean square error (RMSE), Mean absolute error (MAE), execution time, and latency of the proposed model are 30, 12, 1.38 s, and 1.8 s, respectively.