Hybrid Quantum Neural Networks for Early Diagnosis of Multiple Sclerosis using Explorable Quantum AI (VQE and QAOA)

This study proposes a hybrid architecture of Quantum Neural Networks (HQNN) combining quantum computing with classical machine learning to improve the early diagnosis of Multiple Sclerosis (MS). The model integrates simulated multimodal data, including Ultra-High Field MRI (7T), liquid biomarkers (Neurofilament Light Chain - NfL), and genetic data. By utilizing variational algorithms such as VQE and QAOA, the simulated HQNN demonstrates superior performance in capturing complex patterns in high-dimensional spaces, achieving higher accuracy and AUC compared to traditional models like SVM and Random Forest. The research also emphasizes Explainable AI (XAI) to ensure clinical transparency.