Abstract Accurate skin lesion classification in real-world deployment remains difficult when models are trained on a single dataset and must generalize across heterogeneous devices and patient cohorts. Quantum computing can enrich representation learning via variational transformations in high-dimensional Hilbert spaces, yet clinically reliable and domain-robust hybrid quantum dermoscopy remains insufficiently explored. A shift-consistent and explanation-stable distributed hybrid quantum network (SCES-QNet) was developed to examine whether source-only cross-domain generalization, calibration, and attribution consistency could be improved within a unified dermoscopic classification framework. Distributed quantum branches were combined with prediction-consistency and attribution-equivariance regularization under strict source-only training. HAM10000 was used for model development, whereas ISIC 2019 and PAD-UFES-20 were reserved for external evaluation after class-taxonomy harmonization. Discrimination, calibration, and explanation-stability measures were assessed over five independently initialized runs. On HAM10000, near-ceiling performance was achieved (Macro-F1 = 98.4%, Balanced Accuracy = 98.6%, macro-AUROC = 0.998). Under source-only transfer, strong retention was maintained on ISIC 2019 and PAD-UFES-20 (Macro-F1 = 96.2% / 94.8%; macro-AUROC = 0.993 / 0.989). Calibration degradation remained controlled (ECE: 0.007 → 0.014 → 0.017), and explanation stability remained high across domains (ESS ≥ 0.872). Under the evaluated protocol, SCES-QNet was associated with strong discrimination, controlled calibration drift, and comparatively stable explanations across the selected datasets. The findings provide preliminary evidence for stability-aware hybrid quantum learning in dermoscopic classification; however, broader multi-center evaluation, verified cross-archive duplicate auditing, and execution on physical NISQ hardware remain necessary before clinical or hardware-level conclusions can be drawn.
Abstract Accurate skin lesion classification in real-world deployment remains difficult when models are trained on a single dataset and must generalize across heterogeneous devices and patient cohorts. Quantum computing can enrich representation learning via variational transformations in high-dimensional Hilbert spaces, yet clinically reliable and domain-robust hybrid quantum dermoscopy remains insufficiently explored. A shift-consistent and explanation-stable distributed hybrid quantum network (SCES-QNet) was developed to examine whether source-only cross-domain generalization, calibration, and attribution consistency could be improved within a unified dermoscopic classification framework. Distributed quantum branches were combined with prediction-consistency and attribution-equivariance regularization under strict source-only training. HAM10000 was used for model development, whereas ISIC 2019 and PAD-UFES-20 were reserved for external evaluation after class-taxonomy harmonization. Discrimination, calibration, and explanation-stability measures were assessed over five independently initialized runs. On HAM10000, near-ceiling performance was achieved (Macro-F1 = 98.4%, Balanced Accuracy = 98.6%, macro-AUROC = 0.998). Under source-only transfer, strong retention was maintained on ISIC 2019 and PAD-UFES-20 (Macro-F1 = 96.2% / 94.8%; macro-AUROC = 0.993 / 0.989). Calibration degradation remained controlled (ECE: 0.007 → 0.014 → 0.017), and explanation stability remained high across domains (ESS ≥ 0.872). Under the evaluated protocol, SCES-QNet was associated with strong discrimination, controlled calibration drift, and comparatively stable explanations across the selected datasets. The findings provide preliminary evidence for stability-aware hybrid quantum learning in dermoscopic classification; however, broader multi-center evaluation, verified cross-archive duplicate auditing, and execution on physical NISQ hardware remain necessary before clinical or hardware-level conclusions can be drawn.
Skin lesion classification from dermoscopic images remains difficult because diagnostically important lesions may share similar colours, textures, and structural patterns. Fine local details and broader lesion characteristics must therefore be considered together to achieve reliable multi-class classification. This study presents a noise-aware and statistically evaluated dual-branch hybrid quantum-classical framework that combines EfficientNet-B3 features, handcrafted dermatological descriptors, and parameterized quantum circuits. A 6-qubit branch processes micro-level textural information, while a 4-qubit branch represents broader structural characteristics. The resulting quantum features are combined with deep visual representations and handcrafted descriptors based on texture, colour, and lesion complexity. The framework was evaluated using stratified five-fold cross-validation and repeated experiments with different random initialization seeds. On HAM10000, it achieved a mean accuracy of 95.18% ±0.75 %, a mean balanced accuracy of 92.92% ±0.89 %, and a mean macro-F1 score of 0.904±0.009 . A separate training and evaluation experiment on ISIC-2019 produced an accuracy of 93.86% and a macro-F1 score of 0.913, indicating consistent performance across two independently evaluated dermoscopic datasets. The effect of non-ideal quantum operation was also examined by introducing depolarizing noise during evaluation. The model retained accuracies of 94.13% at p=0.01 and 93.21% at p=0.05 . These findings show that combining deep features, dermatological descriptors, and dual-scale quantum representations can improve the discrimination of visually similar lesion categories, including melanoma. The proposed framework provides a statistically reliable and noise-resilient approach for computer-aided dermoscopic image classification, although further external and clinical validation is required before practical clinical use.
This paper presents a systematic ablation and evaluation integrity study of hybrid quantum-classical neural networks for seven-class skin lesion classification on the HAM10000 dermoscopy dataset. We study the effect of the evaluation protocol on a hybrid quantum-classical pipeline by comparing two partitioning schemes on the same model and data. The first is a leakage-prone scheme that partitions the pool of augmented images, and the second is a verified scheme that partitions original lesion identities. Under the leakage-prone scheme, augmented variants of the same physical lesion appear on both sides of the split. An audit found that 1,095 of 4,810 lesions (22.8%) crossed the split boundary, artificially inflating all previously reported metrics. We rebuild the pipeline with a verified lesion-level partition consisting of 4,152 training, 522 validation, and 517 original test images. The training set is augmented to 10,503 images while ensuring zero image-level and lesion-level overlap. We then re-run every experiment across five random seeds with per-class Wilson confidence intervals and paired significance tests on a fixed held-out test set that is never used for tuning. Under the corrected protocol, the strongest configuration features a ResNet50 backbone with concatenation fusion, focal loss, and test-time augmentation. This setup reaches 78.41% ±0.74 test accuracy and a macro F1 of 0.804 against a verified baseline of 76.17% ±1.91 . However, under the leakage-prone protocol, this exact same pipeline yields 94.57%. A parameter-matched classical control, in which a 36-parameter multilayer perceptron replaces the variational quantum circuit (VQC), is statistically indistinguishable from the hybrid model (78.14% versus 78.41%, p = 0.81). Additionally, a controlled fusion comparison shows that concatenation is not significantly better than weighted fusion. We therefore report no detectable advantage from the quantum branch at this scale as a controlled negative result rather than a failure, and we characterize VQC trainability under noiseless and depolarizing simulation. Finally, a split-audit survey of recent quantum-hybrid HAM10000 studies indicates that reported accuracies above 90% are associated with undisclosed or leakage-prone evaluation protocols. The contribution of this work is methodological: a reproducible leakage audit, a verified correction protocol, an honest multi-seed baseline, a parameter-matched classical control, and evidence that evaluation protocols rather than architectures drive the headline differences reported in this literature.
Skin cancer is among the most widespread and life-threatening conditions in the world with the need to detect lesions promptly and consistently in order to treat the disease in a timely manner. Nonetheless, automated lesion classification of the skin is a complex challenge due to inter-class similarity, which is high, illumination variation, and imbalance in the data in various classes of lesions. To address these obstacles, this paper presents a lightweight but powerful 6-qubit Hybrid Quantum-Classical Convolutional Neural Network (HQNN) that is optimized to 6-bit (low-precision) medical imaging applications. The proposed framework will use a ResNet-18 backbone to learn deep dermatological features based on dermoscopic images and a Particle Swarm Optimization (PSO) algorithm with a Random Forest (RF) ranking of features will reduce the 512-dimensional embedding space into six highly discriminative descriptors, which can be encoded in quantum. These are then converted into quantum states with ZZFeatureMap and RealAmplitudes ansatz which comprise a 6-qubit variational circuit the output of which is fused with a classical CNN classifier to achieve final lesion prediction. This strategy of optimization of quantum-aware features is a balance between the representation of high dimensions and limitations of qubits, which ensures a computational tractability on near-term noisy intermediate-scale quantum (NISQ) devices. Existing experimental results on the ISIC 2019 skin-lesion dataset show state-of-the-art performance achieving 98.2% accuracy, 98.5% precision, 97.7% recall, 98.0% F1-score, and a macro ROC-AUC of 0.985. The findings affirm that quantized deep learning with low-qubit quantum encoding has better diagnostic performance, noise resistance, and scalability with minimal hardware costs, making Q-SkinNet a realistic advancement towards experimental quantum-assisted dermatological diagnostics.
Skin cancer identification and categorization are still open issues in medical image analysis due to ambiguous boundaries, variance in shapes, and strong similarities to non-malignant lesions, which must be acknowledged by any framework supporting the clinical decision making. The proposed Quantum-Enhanced Deep Learning Framework (QDLF) is a new approach that combines Quantum feature encoding and Classical deep learning networks for skin lesion classification. Hereby, the QDLF is compared with the traditional models such as ResNet50, DenseNet, and VGG-16 showing a better performance on the HAM10000 dataset containing 10,015 dermoscopic images with seven lesion categories. Therefore, higher accuracy was achieved in the proposed framework at an accuracy level of 96.2% with F1-score 95.1% and ROCAUC value of 0.983 as compared to classical approaches. The QDLF uses quantum feature mappings with Variational Quantum Circuits (VQCs) that enables the model to learn abstract non-linear patterns and it incorporates global contextual features from pre-trained CNNs. The use of a mixture of CNN and FNN in this study not only improves the classification efficiency but also cuts down the time taken to train and the number of parameters involved, getting to a convergence point in 36 minutes with 18.5 million of parameters. To further enhance the interpretability of the model, Grad-CAM visualization is employed to identify clinically significant areas of the lesions as well as t-SNE plots showing good distinction in the quantum features space. The findings affirm the effectiveness of QDLF as a method for addressing class imbalance issues, with precision and recall for key classes such as melanoma at 92.7% and 94.1% respectively. Quantum-classical hybrid frameworks presented in this work reveal the uniqueness of the approach to medical image analysis in terms of scalability and applicability to real-life cases. Further studies will focus on implementation on quantum platforms and generalization of this approach to other image processing tasks in medicine.
BACKGROUND: Accurate classification of skin cancer is critical for early detection and timely treatment, significantly improving patient survival rates. While quantum neural networks combined with transfer learning show promise in medical image analysis, quantum noise remains a major challenge, compromising the stability and reliability of these systems. This study aims to address this limitation by developing a robust quantum-based framework for skin cancer classification. METHODS: We propose a Quantum Dual-Branch Neural Network (QDBNN) that employs two independently trained network branches without shared weights. Dual-modal features are fused at the fully connected layer, and a Variational Quantum Classifier (VQC) is utilized for final classification. The model is evaluated on two datasets: the multiclass HAM10000 and the binary Malignant vs. Benign dataset. RESULTS: QDBNN achieved state-of-the-art accuracies of 93.6% on HAM10000 and 93.5% on the Malignant vs. Benign dataset, outperforming classical and quantum transfer learning baselines. The dual-branch architecture and weighted feature fusion demonstrated enhanced robustness against quantum noise while improving generalization. CONCLUSION: QDBNN effectively mitigates quantum noise interference and leverages quantum-classical hybrid advantages for skin cancer classification. Its success highlights the potential of quantum-inspired architectures in medical imaging, offering a pathway toward clinically deployable tools for early diagnosis. Future work will focus on hardware optimization and scalability to larger datasets.
Achieving reliable radiological image classification requires not only high accuracy, but also well-calibrated uncertainty estimates, robustness in the face of domain shifts, and generalizability across various imaging modalities. Current deep learning methodologies for medical image classification frequently display overconfident predictions and experience significant degradation under realistic clinical conditions, such as label noise, limited training data, and distribution shifts between institutions. Hybrid quantum–classical machine learning has emerged as a promising paradigm; however, existing studies in medical imaging predominantly focus on single-modality and accuracy-centric approaches. In this study, we introduced Q-CaMIR (Cross-Modal, Uncertainty-Aware Hybrid Quantum–Classical Learning), a novel framework that integrates chest X-ray and brain MRI classification within a unified hybrid quantum–classical architecture, incorporating explicit uncertainty quantification. Q-CaMIR utilizes a pretrained classical vision backbone to extract modality-agnostic latent representations, aligns cross-modal features through a contrastive representation module, and channels the resulting compact embeddings through a variational quantum circuit (VQC) decision head, enhanced with evidential deep learning for uncertainty-aware prediction. We evaluated Q-CaMIR on three public benchmarks: CheXpert and MIMIC-CXR (external validation) for chest radiograph classification, and BraTS 2021 for brain tumor classification, employing a comprehensive suite of metrics including accuracy, macro-F1, AUROC, AUPRC, expected calibration error (ECE), and Brier score. Extensive experiments under low-data regimes, synthetic label noise, intensity corruption, and cross-institutional domain shifts demonstrate that Q-CaMIR achieves competitive classification performance compared to robust classical baselines while significantly improving calibration (up to 47% ECE reduction) and robustness. Ablation studies confirm the complementary contributions of the quantum decision head and uncertainty-aware training objective. Our findings suggest that hybrid quantum–classical learning provides a meaningful inductive bias for uncertainty-sensitive radiological classification, advancing towards clinically deployable AI systems for radiation-related diagnostic workflows.