Quantum Physics-Informed Based Support Vector Machine for Cardiovascular Disease Prediction
Mundla Narasimhappa, Saravanan S
Cardiovascular disease (CVD) is a primary cause of death globally, necessitating accurate and physiologically compatible prognostic models. The nonlinear, multi-scale dynamics of cardiovascular systems despite following physical rules are frequently missed by traditional machine learning techniques. We proposed the Quantum Physics-Informed Neural Network-based Support Vector Machine (QPINN-SVM) framework as a solution to this constraint. Three main parts of the approach are integrated: support vector machine classification, quantum kernel generation, and modeling influenced by physics. In order to ensure scientific validity, physics-informed neural networks incorporate arterial blood pressure dynamics, ECG waveform limitations, and hemodynamic equations into the learning process. Using parameterized quantum circuits, these embeddings are transferred into quantum Hilbert space, yielding expressive kernels that capture over 90% of nonlinear interactions in multimodal patient data. In comparison to classical baselines, the suggested approach, which is trained on these quantum kernels, shows a 15-20% increase in predicting accuracy. By integrating clinical decision support, quantum machine learning, and physicsinformed modeling into a single predictive system, the QPINN-SVM paradigm shows promise for pushing healthcare analytics beyond existing approaches.
International Conference on Trends in Quantum Computing and Emerging Business Technologies (TQCEBT)•Apr 1, 2026•0 citations