A brain stroke is a critical cerebrovascular disorder that disrupts blood flow to specific brain regions, leading to irreversible neuronal damage if not diagnosed promptly. Accurate and early classification of stroke subtype plays a vital role in reducing mortality and improving recovery outcomes. Traditional diagnostic methods such as MRI and CT imaging rely heavily on manual interpretation, which can be time-consuming and prone to subjective variability. This research presents a Quantum support vector machine (QSVM) framework for automated brain stroke classification. The proposed system integrates a unified classical feature extraction pipeline comprising textural, morphological, frequency-domain, and statistical descriptors, followed by quantum state encoding within a six-qubit circuit using a ZZFeatureMap-based quantum kernel. The quantum kernel maps classical features into a higher-dimensional Hilbert space to enhance nonlinear separability. Experimental evaluation on a publicly available Kaggle MRI dataset using stratified 5-fold cross-validation demonstrates that the QSVM achieves 96.8% classification accuracy, 96.2% precision, 97.1% recall, F1-score of 96.6%, and an AUC-ROC of 0.982, outperforming optimized classical baselines including Random Forest, K-Nearest Neighbors, and traditional SVM variants on identical feature sets. All experiments were conducted using a classical quantum simulator; therefore, the reported improvements represent simulator-based performance gains rather than hardware-level quantum advantage. These findings suggest that quantum-inspired kernel methods can improve classification performance under controlled experimental conditions, warranting further validation on larger multicenter datasets and real quantum hardware.
A brain stroke is a critical cerebrovascular disorder that disrupts blood flow to specific brain regions, leading to irreversible neuronal damage if not diagnosed promptly. Accurate and early classification of stroke subtype plays a vital role in reducing mortality and improving recovery outcomes. Traditional diagnostic methods such as MRI and CT imaging rely heavily on manual interpretation, which can be time-consuming and prone to subjective variability. This research presents a Quantum support vector machine (QSVM) framework for automated brain stroke classification. The proposed system integrates a unified classical feature extraction pipeline comprising textural, morphological, frequency-domain, and statistical descriptors, followed by quantum state encoding within a six-qubit circuit using a ZZFeatureMap-based quantum kernel. The quantum kernel maps classical features into a higher-dimensional Hilbert space to enhance nonlinear separability. Experimental evaluation on a publicly available Kaggle MRI dataset using stratified 5-fold cross-validation demonstrates that the QSVM achieves 96.8% classification accuracy, 96.2% precision, 97.1% recall, F1-score of 96.6%, and an AUC-ROC of 0.982, outperforming optimized classical baselines including Random Forest, K-Nearest Neighbors, and traditional SVM variants on identical feature sets. All experiments were conducted using a classical quantum simulator; therefore, the reported improvements represent simulator-based performance gains rather than hardware-level quantum advantage. These findings suggest that quantum-inspired kernel methods can improve classification performance under controlled experimental conditions, warranting further validation on larger multicenter datasets and real quantum hardware.
The prompt detection of neurodegenerative diseases (NDDs) like Alzheimer's Disease (AD), Mild Cognitive Impairment (MCI), and Parkinson's Disease (PD) is essential yet difficult. This research proposes a hybrid quantum-classical machine learning methodology to categorize NDD using multimodal brain imaging data.ADNI and PPMI datasets' MRI, PET, and fMRI images were preprocessing by Python-based tools such as Scikit-image, SimpleITK, and FreeSurfer via Nipype. Major steps were denoising, registration, normalization, segmentation, and extraction of features from structural and functional imaging. Training was done with a Quantum Support Vector Machine (QSVM) using Qiskit's ZZFeatureMap and QuantumKernel and compared with a conventional SVM. Performance measures-precision, recall, F1-score, class accuracy, AUC, and 10-fold cross-validation-were calculated. QSVM performed better than the standard SVM on all metrics, with 91.25% overall accuracy and macro F1-score of 0.913 as opposed to 87.75% and 0.879 for the SVM. The results validate QSVM's advantage in handling complex, high-dimensional neuroimaging data and its relevance to aiding clinical diagnosis. Keywords: Quantum Machine Learning, Neuroimaging, QSVM, Alzheimer’s Disease, Parkinson’s Disease, Multimodal Classification
Background Lung cancer is a leading cause of cancer mortality, and early detection is critical for patient outcomes. Support Vector Machines (SVM) have been widely applied to diagnostic classification, but their performance can be limited in terms of sensitivity. Quantum Support Vector Machines (QSVM) offer a potential approach to improve this aspect. Methods We compared SVM and QSVM on balanced lung cancer subsets (1,000 samples each) using k-fold cross-validation. QSVM employed ZZFeatureMap and FidelityQuantumKernel for quantum feature encoding. Results QSVM achieved an average recall improvement of about 8% over classical SVM, while maintaining comparable precision and accuracy. Under a realistic 1:100 class imbalance, precision declined, highlighting challenges with highly skewed data. Conclusion QSVM shows promise for enhancing sensitivity in medical diagnostics but requires further validation on larger, heterogeneous datasets and real quantum hardware. Clinical trial number Not applicable.
Purpose This study investigates the integration of quantum-enhanced machine learning (QML) into medical image analysis, focusing on classification and feature extraction tasks. It aims to evaluate whether hybrid quantum-classical models can overcome the limitations of traditional machine learning in handling high-dimensional, clinically relevant imaging data. Design/methodology/approach Hybrid architectures combining classical preprocessing with quantum components, specifically variational quantum circuits (VQCs) and quantum kernel methods, were developed and implemented in simulated quantum environments using Qiskit and PennyLane on Google Colab. Publicly available datasets (ChestX-ray14, Brain Tumor MRI and CT Hemorrhage) were preprocessed and encoded into quantum states. Model performance was evaluated against classical baselines (MLP and CNN) using accuracy, precision, recall, F1-score, convergence speed and trainable parameters. Findings Quantum-enhanced models achieved competitive or superior performance compared to classical models with significantly fewer trainable parameters and faster convergence. The Quantum Kernel SVM demonstrated the highest precision (89.1%) and F1-score (88.5%) using only 8 parameters, while VQC models converged up to 2× faster than classical baselines. These results highlight the potential of QML to improve efficiency, robustness and scalability in diagnostic imaging workflows, particularly in resource-constrained or data-limited settings. Research limitations/implications The study relies on simulated quantum environments, which do not capture hardware noise or qubit limitations. Future work will extend to real quantum devices, multimodal data integration and interpretability strategies. Practical implications Quantum kernel methods and lightweight VQC architectures enable faster, resource-efficient diagnostic modeling, making them suitable for clinical environments with limited computational capacity or time-sensitive decision-making needs. Social implications Improving the efficiency and accessibility of medical image analysis has the potential to support faster diagnosis and enhance healthcare delivery, particularly in underserved or resource-limited regions. Originality/value This study provides a practical evaluation of hybrid quantum-classical models applied to multiple real medical imaging datasets in a reproducible simulated environment. It offers an early-stage framework demonstrating how QML techniques can be integrated into existing machine learning pipelines for medical image analysis.
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.
Heart disease remains a leading cause of mortality globally, driving interest in advanced computational methods forearly prediction. This study presents a comparative analysis of classical machine learning techniques versus a QuantumSupport Vector Machine (QSVM) for predicting heart disease using a specialized dataset featuring standard clinicalmetrics alongside a unique 'QuantumPatternFeature. 'We implemented and evaluated six classical algorithms (LogisticRegression, K-Nearest Neighbors, Decision Tree, Random Forest, Naive Bayes, Gradient Boosting), employingRandomizedSearchCV for hyperparameter optimization. For the quantum approach, a QSVC utilizing Qiskit'sZZFeatureMap (reps=2) and FidelityQuantumKernel was trained and tested using a classical simulator. A classicalSupport Vector Classifier (SVC) with an RBF kernel served as an additional baseline. Performance was assessed on a20% hold-out test set using accuracy, precision, recall, F1-score, and confusion matrices. Results indicate that tunedclassical models, particularly Logistic Regression and Gradient Boosting, achieved the highest accuracy (93.0%),significantly outperforming the simulated QSVC (79.0%). The classical SVC baseline also surpassed the QSVM with89.0% accuracy. Furthermore, the QSVM exhibited substantially higher computational times for both training andprediction on the simulator compared to all classical methods. The findings suggest that for this dataset size and thespecific simulated QSVM configuration, classical machine learning approaches offer superior performance andefficiency, and the 'QuantumPatternFeature' did not confer a discernible advantage within this quantum implementationframework.
Multiclass brain tumor classification from magnetic resonance imaging (MRI) remains challenging due to subtle inter-class visual similarities, particularly for anatomically small and heterogeneous tumor types such as pituitary lesions. To address these challenges, this paper presents a hybrid classical-quantum learning framework that integrates a deep convolutional feature extractor from the ResNet family with a variational quantum neural network (QNN) head for enhanced feature representation. In the present work, classical deep features extracted by the ResNet model are encoded into a six-qubit quantum circuit, enabling richer feature interactions within a high-dimensional Hilbert space. In addition, an anatomically guided Kernel-based Center Cropping (KCC) strategy is introduced as a deterministic spatial prior to emphasize anatomically relevant central brain structures, particularly benefiting challenging pituitary tumor cases. The proposed Kernel-based Center Cropping–Hybrid Quantum Neural Network (KCC-HQNN) framework is trained and evaluated on a publicly available composite four-class brain MRI dataset curated from Kaggle sources, comprising glioma, meningioma, pituitary tumor, and no-tumor categories. Experimental results across multiple ResNet variants demonstrate that the hybrid quantum–classical models consistently outperform their purely classical counterparts, achieving classification accuracy of up to 99%, while macro-averaged precision, recall, and F1 scores consistently fall in the range of approximately 0.95–0.99 across different ResNet variants. Reliability and reproducibility are further validated through stratified five-fold cross-validation and repeated experiments using multiple random initialization seeds, yielding a mean classification accuracy of 99.00 ± 0.08%, macro F1-score of 0.986 ± 0.11, and weighted F1-score of 0.989 ± 0.12, while cross-dataset evaluation on independent IEEE DataPort MRI datasets demonstrates robust generalization under dataset shift. All experiments are conducted on publicly available retrospective datasets using a quantum simulator. These results indicate that combining quantum feature processing with anatomically informed spatial processing improves multiclass discrimination of brain tumors.