IEEE International Conference on Healthcare Informatics (ICHI)
A Hybrid Classical-Quantum Machine Learning Framework for Tumor Mutation Classification: A Comprehensive Study of Quantum Optimizers in Medical Image Analysis
Md Jobair Hossain Faruk, Md Kamrul Siam, Jerry Q. Cheng, Huanying Gu, Louis P. Dehner, Mai He
Classical machine learning methods face scalability and optimization challenges when applied to large-scale histopathological datasets. Quantum machine learning (QML) offers a promising direction by enabling compact high-dimensional feature encoding and alternative optimization strategies. In neuroblastoma (NB) histopathology, where accurate tumor mutation classification is critical for clinical decision-making, such capabilities are essential to capture subtle molecular variations often overlooked by conventional models. This study presents a hybrid classical-quantum machine learning framework for tumor mutation classification, combining ViT embeddings with a 4-qubit variational quantum circuit. We conducted experiments using 594,271 tumor relevant patches of size 512 × 512 pixels, collected from 97 manually annotated Whole Slide Images (WSIs) (≤ 150,000 × 100,000 pixels) provided by domain experts which also represents 85 NB cases across three U.S. medical institutions. The classical Adam baseline achieved 91.8% accuracy with tumor sensitivity of 97.3%. The hybrid model with Simultaneous Perturbation Stochastic Approximation (SPSA) reached 91.0% accuracy, while Quantum Natural Gradient (QNG) and Rotoselect underperformed (89.8% and 81.3%, respectively). Ablation experiments showed that both Quantum Multi-Head Attention (QMHA) and Quantum MLP (QMLP) blocks provide complementary gains, with circuit depth L=2 and vector readout offering the best trade-off between expressivity and stability. We further evaluated computational performance across dataset scales. Classical ViT exhibited a linear growth of approximately 21.33 minutes per WSI, while the hybrid QML model scaled more gradually at 19.36 minutes per WSI. Although QML incurred higher overhead at smaller dataset sizes, its slower growth rate allowed it to require less processing time than ViT starting at ≈ 133 WSIs, underscoring its potential for improved scalability on large datasets. Overall, this work establishes a systematic benchmark of hybrid QML for digital pathology, demonstrating competitive classification performance and the potential for improved efficiency as quantum hardware matures.
Classical machine learning methods face scalability and optimization challenges when applied to large-scale histopathological datasets. Quantum machine learning (QML) offers a promising direction by enabling compact high-dimensional feature encoding and alternative optimization strategies. In neuroblastoma (NB) histopathology, where accurate tumor mutation classification is critical for clinical decision-making, such capabilities are essential to capture subtle molecular variations often overlooked by conventional models. This study presents a hybrid classical-quantum machine learning framework for tumor mutation classification, combining ViT embeddings with a 4-qubit variational quantum circuit. We conducted experiments using 594,271 tumor relevant patches of size 512 × 512 pixels, collected from 97 manually annotated Whole Slide Images (WSIs) (≤ 150,000 × 100,000 pixels) provided by domain experts which also represents 85 NB cases across three U.S. medical institutions. The classical Adam baseline achieved 91.8% accuracy with tumor sensitivity of 97.3%. The hybrid model with Simultaneous Perturbation Stochastic Approximation (SPSA) reached 91.0% accuracy, while Quantum Natural Gradient (QNG) and Rotoselect underperformed (89.8% and 81.3%, respectively). Ablation experiments showed that both Quantum Multi-Head Attention (QMHA) and Quantum MLP (QMLP) blocks provide complementary gains, with circuit depth L=2 and vector readout offering the best trade-off between expressivity and stability. We further evaluated computational performance across dataset scales. Classical ViT exhibited a linear growth of approximately 21.33 minutes per WSI, while the hybrid QML model scaled more gradually at 19.36 minutes per WSI. Although QML incurred higher overhead at smaller dataset sizes, its slower growth rate allowed it to require less processing time than ViT starting at ≈ 133 WSIs, underscoring its potential for improved scalability on large datasets. Overall, this work establishes a systematic benchmark of hybrid QML for digital pathology, demonstrating competitive classification performance and the potential for improved efficiency as quantum hardware matures.
Accurate and computationally efficient classification of brain tumors from MRI scans is critical for deployment in resource-constrained clinical settings, where conventional deep CNN architectures-despite achieving strong accuracy-often require large parameter counts that limit scalability. This work proposes a hybrid quantum-classical deep learning framework that combines a classical convolutional feature extractor with a variational quantum circuit (VQC) for multi-class tumor classification. Spatial features extracted by the CNN backbone will be dimensionality-reduced and encoded into quantum states using angle encoding; a parameterized quantum circuit will then process these features in a higher-dimensional Hilbert space, leveraging superposition and entanglement to capture non-linear feature correlations. Measurement outputs from the quantum layer will be passed to a classical fully connected layer for classification into glioma, meningioma, pituitary tumor, and no-tumor categories. The proposed framework will be evaluated on a publicly available benchmark MRI dataset and benchmarked against state-of-the-art classical CNN architectures in terms of accuracy, precision, recall, F1-score, and parameter efficiency. This study aims to demonstrate that quantum-enhanced layers can achieve competitive classification performance while substantially reducing trainable parameters, offering a practical direction for efficient, quantum-assisted medical image analysis.
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.
Abstract The convergence of quantum computing and medical image analysis has emerged as a promising frontier for advancing diagnostic intelligence beyond the limits of classical machine learning. Recent developments in quantum hardware and quantum machine learning (QML) algorithms have stimulated growing interest in their potential to enhance complex medical imaging tasks, including disease classification, image segmentation, and clinical decision support (Wei et al., 2023; Ullah and Garcia-Zapirain, 2024). This study presents a comprehensive review and synthesis of state-of-the-art QML methodologies applied to medical imaging, systematically examining both theoretical foundations and empirical implementations. We analyze quantum-based and hybrid quantum–classical models employed across modalities such as MRI, CT, X-ray, and histopathological imaging, with particular attention to classification accuracy, computational efficiency, and diagnostic robustness (Maheshwari et al., 2022; ElBedoui et al., 2025). Our findings reveal clear trends indicating that hybrid quantum–classical approaches currently dominate practical applications, consistently outperforming purely classical baselines in select low-dimensional and domain-specific diagnostic tasks. Performance gains are most evident in feature-rich yet data-constrained scenarios, where quantum kernels and variational circuits demonstrate enhanced representational capacity (Ajlouni et al., 2023; Landman et al., 2022). However, these advantages remain context-dependent and sensitive to noise, circuit depth, and encoding strategies. Overall, QML holds significant promise for improving medical image-based diagnosis, but its widespread clinical adoption is constrained by hardware scalability, limited qubit coherence, and unresolved data encoding challenges. Addressing these limitations will be critical for translating QML from experimental feasibility to routine medical practice (Marengo and Santamato, 2025; Rawas and AlSaeed, 2026).
Quantum machine learning (QML) presents a promising avenue for addressing complex classification challenges, yet its application in medical imaging remains largely unexplored. This work introduces a hybrid quantum classical framework designed to classify skin diseases—Chickenpox, Measles, Monkeypox, and Normal—using a 4 qubit quantum system. The framework integrates a Quantum Convolutional Neural Network (QCNN) for feature extraction and a Variational Quantum Classifier (VQC) for classification, processing a dataset of 770 images with significant class imbalance. To address this imbalance, we employ a class weighted loss function with weightswyi= [3.0, 3.5, 0.6, 0.4], ensuring balanced representation across all classes. The model is optimized using Simultaneous Perturbation Stochastic Approximation (SPSA) over 150 iterations, achieving a test accuracy of 70.13% on a subset of 154 images, which represents 20% of the dataset. This marks a substantial improvement over both a random baseline of 25% and a prior run that achieved 37.66% accuracy, which was skewed toward the majority class. Principal Component Analysis (PCA) reduces image features to a 4 dimensional representation, enabling compatibility with the quantum system. The QCNN extracts these features, while the VQC performs classification based on measurements from 1024 shots on the QASM simulator. To ensure practical reusability, the trained model—including quantum circuits, optimized parameters, and PCA transformations—is saved for future deployment. Validation on a separate set of 24 samples (6 per class) yields 70.83% accuracy (17 out of 24 correct predictions), demonstrating the model’s generalization capabilities. While the performance falls short of classical benchmarks, this study establishes the feasibility of QML with minimal qubits, paving the way for quantum driven diagnostics in resource constrained environments.
Brain tumor classification remains a challenging task due to the complex heterogeneity of glioma characteristics and the limitations of conventional machine learning approaches in effectively integrating multimodal biomedical data. Recent advances in quantum machine learning have demonstrated promising capabilities for handling high-dimensional data; however, existing models often suffer from limited feature representation, insufficient multimodal integration, and optimization challenges. This paper proposes an Enhanced Hybrid Quantum–Classical Architecture (EHQCA) for glioma classification by combining MRI imaging features and TCGA molecular biomarkers within a unified learning framework. The proposed methodology employs ResNet-50 for imaging feature extraction, ensemble feature selection for molecular biomarkers, attention-based multimodal fusion, adaptive amplitude encoding, and a Multi-Layer Variational Quantum Classifier (ML-VQC). A hybrid Adam-SPSA optimization strategy is introduced to improve convergence speed and training stability. Experimental analysis demonstrates that the proposed EHQCA framework achieves superior classification performance, attaining 91.2% accuracy while reducing training epochs compared with existing quantum and classical models. The results highlight the effectiveness of multimodal fusion and quantum-enhanced learning for improving brain tumor diagnosis and supporting precision medicine applications
Medical image classification is a critical task in computer-aided diagnostics, but classical deep learning models often demand substantial computational resources and large parameter spaces. This paper investigates the potential of hybrid quantum-classical neural networks to address this challenge in the context of multi-class brain tumor classification from MRI scans. We introduce a custom hybrid quantum machine learning (QML) model and conduct a rigorous comparative analysis against its purely classical counterpart, as well as established architectures such as ResNet-18 and VGG-16. Our findings demonstrate that the proposed hybrid QML model achieves a state-of-the-art test accuracy of 96.58%, matching the performance of the significantly larger ResNet-18 model. Crucially, the hybrid model outperforms its direct classical counterpart, the Classical Benchmark (96.20%), while utilizing approximately 80% fewer parameters. Although simulating quantum circuits results in longer training times, our findings highlight the remarkable parameter efficiency and competitive accuracy of hybrid quantum models, suggesting a promising direction for developing powerful yet lightweight networks for medical diagnostics.