Advancing Healthcare Diagnostics with a Hybrid AI Model

Abstract

Artificial intelligence (AI) has significantly impacted the field of healthcare by advancing medical diagnostics and decision-making processes. This paper presents a unified framework that integrates Convolutional Multi-Layer Perceptrons (Conv MLP), Convolutional Neural Networks (ConvNet), and Vision Transformers (ViTs) to address two critical tasks: sentiment analysis of medical reports and medical image recognition. By leveraging the distinct strengths of each architecture, the proposed hybrid model improves performance, effectively balancing computational efficiency, localized feature extraction, and global dependency modeling. The findings demonstrate that the model reduces task-specific errors and enhances convergence rates, achieving superior results in sentiment and image recognition tasks. This study advances the field of medical AI by overcoming the limitations of single-architecture models and establishing the feasibility of hybrid systems for robust, multi-modal applications.