Quantum machine learning (QML) offers a promising framework for analyzing complex, high-dimensional biomedical data, yet rigorous benchmarking on real-world clinical datasets remains limited. In this study, we evaluate a quantum support vector machine (QSVM) for agitation detection using the Technology Integrated Health Management (TIHM) smart-home dementia dataset. The proposed QSVM uses a 13-qubit fidelity-based feature map with two degrees of freedom in angle encoding to process the same 26 features used by the strongest classical baselines under a subject-independent evaluation protocol. Across seven ID-disjoint trials, the QSVM achieves a weighted F1-score of$0.92 \pm 0.06$, with consistently high precision of$0.93 \pm 0.06$and recall of$0.91 \pm 0.06$. These results demonstrate that quantum kernel methods can match or exceed state-of-the-art classical performance while providing a reproducible benchmark for cognitive-health prediction. This work lays the foundation for future exploration of hybrid quantum–classical models and clinical decision-support applications in behavioral health monitoring.
Quantum machine learning (QML) offers a promising framework for analyzing complex, high-dimensional biomedical data, yet rigorous benchmarking on real-world clinical datasets remains limited. In this study, we evaluate a quantum support vector machine (QSVM) for agitation detection using the Technology Integrated Health Management (TIHM) smart-home dementia dataset. The proposed QSVM uses a 13-qubit fidelity-based feature map with two degrees of freedom in angle encoding to process the same 26 features used by the strongest classical baselines under a subject-independent evaluation protocol. Across seven ID-disjoint trials, the QSVM achieves a weighted F1-score of0.92±0.06, with consistently high precision of0.93±0.06and recall of0.91±0.06. These results demonstrate that quantum kernel methods can match or exceed state-of-the-art classical performance while providing a reproducible benchmark for cognitive-health prediction. This work lays the foundation for future exploration of hybrid quantum–classical models and clinical decision-support applications in behavioral health monitoring.
Timely clinical intervention at the early stage for detecting widespread psychological health issues such as depression and anxiety is essential. Conventional diagnostic approaches that are generally based on subjective questionnaires face limitations which can result in delays as well as imprecisions. Current data-driven methodologies employ features that are extracted from multiple sources (behavioural surveys, clinical attributes and physiological signals excluding EEG and ECG). Although known classical ML models such as RF and LR are effective, the limitations of these models on capturing extremely complex non-linear issues result in their performance being limited. This performance upper bound is validated by the combined classical benchmark accuracy of 86.59%. This paper introduces a new quantum hybrid support vector machine (QSVM) framework, and shows how it can dramatically improve prediction accuracy on such mental health conditions. The system uses a hybrid design, in which the classical processing part computes and sends out a processed feature map to be quantum mapped into a Q exponentially larger-dimensional Hilbert space through quantum kernel methods (Quantum Feature Maps). By taking advantage of the quantum effects, the QSVM implicitly computes kernels that are difficult for classical machines and allows achieving better classification boundaries. The accuracy of QSVM is critically evaluated against the RF and LR benchmarks. Thus The Hybrid QSVM shows preliminary signs of quantum advantage, particularly in precision and kernel separability, although accuracy remains comparable to classical baselines. Validation on real quantum hardware will be required to establish practical quantum advantage.
Early and accurate anomaly detection in Intensive Care Unit (ICU) monitoring is vital for improving patient outcomes and reducing mortality. Traditional machine learning techniques often face challenges in handling the noisy, high-dimensional, and limited datasets characteristic of ICU physiological signals. In this study, we present an early investigation of a quantum-enhanced framework employing a classically simulated Quantum Support Vector Machine (QSVM) to improve the detection of critical anomalies in real-world ICU time-series data. Utilizing the publicly available PhysioNet 2012 Challenge dataset, we extract comprehensive statistical and temporal features from vital signs including heart rate, blood pressure, respiratory rate, and oxygen saturation. Our QSVM model leverages a quantum kernel mapped via Qiskit’s ZZFeatureMap and is rigorously benchmarked against classical machine learning classifiers such as Support Vector Machines with radial basis function kernels, Random Forest, and XGBoost. Experimental results demonstrate that the QSVM shows competitive and robust classification performance, particularly excelling in scenarios with limited and noisy data, with an AUC of 0.92, particularly excelling in scenarios with limited and noisy data. Furthermore, the quantum kernel’s ability to implicitly project data into a high-dimensional Hilbert space provides enhanced separability of complex physiological patterns. This work constitutes one of the first simulation-based applications of quantum machine learning to critical care anomaly detection, underscoring the promise of hybrid quantum-classical approaches for advancing real-time clinical decision support systems.
Early detection of Autism Spectrum Disorder (ASD) is crucial for initiating interventions that can significantly improve developmental outcomes. Traditional diagnostic methods are often subjective and time-consuming, leading to delays in diagnosis. While classical machine learning models have been applied to ASD detection with promising results, they face challenges in processing complex, high-dimensional medical data. Quantum computing utilizes superposition, and entanglement offers a novel approach to handle such data more efficiently. This study investigates the application of Quantum Support Vector Machines (QSVM) for ASD detection using the Autism Screening Data for Toddlers (ASDTests) dataset. After preprocessing and feature selection, we trained a QSVM model. We compared its performance with classical machine learning models, including Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbors (KNN), and Naive Bayes (NB). We evaluated the models based on accuracy, precision, recall, and F1-score across multiple random subsets to assess robustness. The QSVM model achieved the highest mean accuracy (96.7 %), precision (97.2 %), and F1-score (95.9 %) with low variance, indicating strong and consistent performance. Statistical analyses using the Paired t-test and Friedman Test showed marginal improvements over the SVM model (p = 0.0495), with no significant difference compared to RF and NB models. These results suggest that QSVM can enhance the accuracy and robustness of ASD detection compared to classical models. Future work could explore hybrid models combining quantum and classical machine learning approaches to improve performance. Applying these methods to different ASD datasets may provide broader validation and insights into their generalizability.
Blood glucose is one of the key indicators influencing the body’s metabolic performance. Accurate prediction and proactive management of this parameter are of paramount clinical importance. In this context, the analysis and evaluation of quantum machine learning (QML) algorithms for processing medical data and enhancing the predictive accuracy of health-related models have emerged as cutting-edge research directions in recent studies. Leveraging the principles of quantum physics and computation, QML algorithms serve as a promising alternative to classical methods for predicting chronic diseases. By mapping classical data into a quantum feature space and exploiting intrinsic quantum properties such as superposition and entanglement, these algorithms significantly enhance the ability to detect complex and latent patterns within datasets. In this study, the dataset was first preprocessed using Principal Component Analysis (PCA) for dimensionality reduction and noise suppression. Subsequently, the predictive performance of two quantum machine learning models—the Quantum Support Vector Machine (QSVM) and the Variational Quantum Circuit (VQC)—was evaluated and compared with that of the classical Support Vector Machine (SVM) for blood glucose level prediction. The results indicate that the QSVM model achieves a prediction accuracy of 90%, outperforming the classical SVM, which attained an accuracy of 74.4%. Moreover, the hybrid classical–quantum VQC model demonstrated the highest predictive performance, achieving an accuracy of 94.2%, thereby establishing it as the most effective approach for blood glucose level forecasting in this experimental setting.
Parkinson’s disease (PD) is a progressive neurodegenerative disorder affecting millions worldwide, where early and accurate diagnosis plays a critical role in improving patient outcomes. While classical machine learning methods have shown strong performance in PD classification using clinical and acoustic data, they remain limited in capturing complex nonlinear feature interactions in high-dimensional spaces. This study explores the application of quantum machine learning (QML) techniques for PD prediction, focusing on Variational Quantum Classifiers (VQC) and Quantum Support Vector Machines (QSVM), and compares their performance with strong classical baselines including Logistic Regression, SVM, Random Forest, Grid Search–optimized models, and XGBoost. Experiments were conducted using stratified train–test splits and validated through cross-validation and statistical testing. Results show that classical models achieve competitive performance, with XGBoost reaching 91.9% accuracy. However, quantum models demonstrate slightly improved performance, with the VQC using EfficientSU2 achieving the highest accuracy of 93.1%, followed by the Real Amplitudes VQC (92.6%) and QSVM (92.2%). These improvements are consistent across evaluation metrics including F1-score and recall. Statistical analysis using 10-fold cross-validation indicates that the VQC (EfficientSU2) shows a significant improvement over classical baselines ( p = 0.031), while other quantum models do not achieve statistical significance over XGBoost ( p > 0.05). McNemar’s test on the independent test set confirms no significant difference in final prediction distributions between quantum and classical models. Overall, the findings suggest that quantum machine learning models are competitive with state-of-the-art classical approaches, offering marginal but consistent performance gains. This study highlights the potential of quantum-enhanced learning frameworks for biomedical classification tasks, while also emphasizing that their advantages remain incremental rather than definitive under current simulation-based implementations.
Cardiovascular disease (CVD) is the leading cause of global mortality, with over 19.8 million deaths recorded in 2022, underscoring the urgent necessity for precise early-stage computational diagnostic techniques. The traditional machine learning approaches tend to fail in the nonlinear feature interactions of high-dimensional clinical data. This paper provides a Quantum Support Vector Machine (QSVM) framework that uses a fidelity-based quantum kernel, calculated using PennyLane Lightning, and a qubit simulator to map patient data into a high-dimensional Hilbert space, enabling complex feature interactions that are unattainable with traditional kernels. The model was tested on a publicly available UCI-based heart disease dataset that included 1,025 patient records and 13 clinical features, such as chest pain type, ST depression, maximum heart rate, and thalassemia test results, with a near-balanced distribution between the classes (51.3% positive, 48.7% negative). All 13 features were represented in quantum states with a Hadamard-RY feature map applied to 13 qubits, and quantum pipelining was used to perform batch executions of the feature map and reduce simulation overhead. The QSVM achieved 93.90% accuracy, outperforming the classical SVM baseline, which achieved 92.50% on the same dataset. Complexity analysis shows O(d) depth, linear qubit scaling, and NISQ-compatible scalable quantum biomedical diagnostics.