QuCardio: Application of Quantum Machine Learning for Detection of Cardiovascular Diseases

Abstract

This research is the first of its kind to leverage the power of Quantum Machine Learning (QML) to perform multi-class classification of Cardiovascular Diseases (CVDs). We propose a novel approach that enables multi-class classification with Pegasos Quantum Support Vector Classifier (QSVC). The QSVC and the Pegasos QSVC significantly outperform the classical SVC by a margin of +10.76% and +9.72%, respectively. The paper further ventures into a quantum deep learning based architecture with a novel Quanvolutional Neural Network (QNN) implementation, outperforming not only its classical CNN counterpart by +3.88% but also the other models by achieving 97.31% accuracy, 97.41% precision, 97.31% recall, 97.30% F1 score, and 99.10% specificity.

Keywords

• Quantum machine learning (QML)
• quantum support vector machine (QSVM)
• Pegasos
• quanvolutional neural network (QNN)
• medical image classification
• cardiovascular disease classification