Machine Learning–Based Classification Framework for Human Health Care Monitoring

Abstract

The healthcare sector generates extensive patient and disease-related data, which, when integrated with artificial intelligence (AI), enables the discovery of latent disease patterns, personalized treatment planning, and condition prediction. Facial expression recognition is a key component in patient health assessment, supporting rapid emergency response. However, variations in posture, scale, occlusion, and illumination often degrade recognition accuracy. This work presents an enhanced facial expression recognition framework employing Local Binary Patterns (LBP) for robust feature extraction, coupled with Discrete Cosine Transformation (DCT) for frequency-domain representation. Three classifiers—Support Vector Machine (SVM), Recurrent Neural Network (RNN), and Naïve Bayes (NB)—are evaluated for integration into wearable health monitoring systems. Experimental validation on standard benchmark datasets demonstrates that the proposed method achieves superior recognition accuracy compared to existing approaches, offering improved reliability for real-time healthcare monitoring applications.