A Hybrid CNN–Vision Transformer and Variational Quantum Circuit-Based Methodology for Noncommunicable Disease Diagnosis and Risk Prediction

Noncommunicable diseases (NCDs) such as cardiovascular diseases, diabetes, cancer and chronic respiratory diseases are the primary causes of death globally. To enhance patient survival and lower health-care burden, it is essential that patient diagnosis is early and that the risk is accurately predicted. Classical machine learning and deep learning algorithms have proven to be successful in healthcare analytics, but they suffer from problems like high dimensionality, multimodality, poor generalization, and lack of interpretability of the results. In this paper, a novel quantum–classical diagnostic framework for NCD prediction and classification is proposed to overcome these drawbacks. The proposed model combines the classical deep learning feature extraction with the quantum enhanced optimization and classification unit. The framework adopts multiple modalities of healthcare data such as medical imaging data, clinical parameters, and genomic information. Experimental results show that the proposed approach yields better classification accuracy, lower dimensionality of extracted features and more efficient learning than the traditional AI methods. The proposed hybrid architecture has the potential to create a scalable and intelligent healthcare solution for next-generation disease diagnosis and personalized risk prediction.