QCNN-Based Diagnostic System for Wilson Disease Using Biochemical and Genetic Markers

Abstract

Wilson disease is a rare hereditary disorder, but it is rising rapidly in recent years. Its silent features results delay in diagnosis, which makes it a more deadly disease. For instance, in Germany, it has increased from 17.8 to 24.4 per million, and South Korea also sees a rise in wave in Wilson disease. Persons suffering from this disease have no initial symptoms but when detected at a later stage, it is irreversible due to the accumulation of copper in the body. Early diagnosis in healthcare is important for effective treatment and to increase survival rates. Conventional diagnostic approaches which biochemical and clinical Tests, can be time-consuming and prone to misinterpretation in early stages. Integration of ML, Neural Networks and Deep Learning Techniques with medical sciences have significantly enhanced the accuracy, efficiency, and reliability of disease diagnosis.This work presents a novel framework based on Quantum ML and Quantum Convolutional Neural Network (QCNN), that combines the properties of quantum computing and CNN to diagnose the disease and improve survival rates. The proposed model leverages quantum feature encoding, parameterized quantum gates, and entanglement-based convolution operations to capture underlying feature relationships more effectively and produces classification outputs through quantum measurement. The experimental analysis is conducted on a publicly available Kaggle dataset having 60,000 records. The experimental results demonstrate that the proposed QCNN model achieves the highest accuracy of 95.30% as compared to classical ML and CNN models. This highlights the potential of proposed approach for early disease diagnosis in real-time clinical applications.