Medical image denoising remains challenging because imaging modalities and acquisition conditions produce heterogeneous and non-stationary noise. Although recent deep-learning and diffusion-based approaches achieve strong restoration performance, many methods rely on fixed transformations or require substantial computational resources. This work proposes the Noise-Adaptive Quantum Attention Network (NAQAN), a hybrid quantum–classical framework for noise-conditioned medical image denoising. NAQAN combines a lightweight convolutional encoder–decoder with an eight-qubit variational quantum attention mechanism whose circuit parameters are dynamically modulated by a specified noise-level conditioning variable. Feature-wise Linear Modulation (FiLM) provides complementary classical conditioning, while residual learning is used for image reconstruction. The present implementation is noise-conditioned rather than fully blind; the corresponding noise level is supplied to the model during both training and evaluation. Experiments on RetinaMNIST, OCTMNIST, Diabetic Retinopathy, and BraTS MRI under Gaussian, speckle, and Poisson corruption demonstrate consistent restoration performance across different imaging modalities and noise conditions. On BraTS MRI, NAQAN achieves a PSNR of 39.11dB and an SSIM of 0.9548. Controlled comparisons with classical denoising baselines and a parameter-matched classical attention variant further evaluate the contribution of the proposed quantum attention mechanism. NAQAN performs denoising through a single forward inference pass; however, simulation of the quantum circuit introduces additional computational overhead relative to the matched classical variant. These results demonstrate the potential of noise-conditioned hybrid quantum–classical attention for medical image restoration while motivating further evaluation on physical quantum hardware and clinically oriented downstream tasks.
Medical image denoising remains challenging because imaging modalities and acquisition conditions produce heterogeneous and non-stationary noise. Although recent deep-learning and diffusion-based approaches achieve strong restoration performance, many methods rely on fixed transformations or require substantial computational resources. This work proposes the Noise-Adaptive Quantum Attention Network (NAQAN), a hybrid quantum–classical framework for noise-conditioned medical image denoising. NAQAN combines a lightweight convolutional encoder–decoder with an eight-qubit variational quantum attention mechanism whose circuit parameters are dynamically modulated by a specified noise-level conditioning variable. Feature-wise Linear Modulation (FiLM) provides complementary classical conditioning, while residual learning is used for image reconstruction. The present implementation is noise-conditioned rather than fully blind; the corresponding noise level is supplied to the model during both training and evaluation. Experiments on RetinaMNIST, OCTMNIST, Diabetic Retinopathy, and BraTS MRI under Gaussian, speckle, and Poisson corruption demonstrate consistent restoration performance across different imaging modalities and noise conditions. On BraTS MRI, NAQAN achieves a PSNR of 39.11dB and an SSIM of 0.9548. Controlled comparisons with classical denoising baselines and a parameter-matched classical attention variant further evaluate the contribution of the proposed quantum attention mechanism. NAQAN performs denoising through a single forward inference pass; however, simulation of the quantum circuit introduces additional computational overhead relative to the matched classical variant. These results demonstrate the potential of noise-conditioned hybrid quantum–classical attention for medical image restoration while motivating further evaluation on physical quantum hardware and clinically oriented downstream tasks.
Medical image classification is a critical component of modern healthcare; however, accurate diagnosis remains challenging due to limited annotated datasets, class imbalance, and the high dimensionality of medical imaging data. To address these challenges, a hybrid quantum-classical neural network (HQCNN) is proposed, integrating classical deep learning with variational quantum learning for medical image classification. The proposed architecture combines a five-layer convolutional neural network (CNN) for hierarchical feature extraction with a lightweight 4-qubit variational quantum circuit (VQC) incorporating quantum state encoding, superposition and entanglement mechanisms, and a quantum attention-Fourier (QAF) module. This hybrid design aims to improve nonlinear feature representation and quantum parameter efficiency while maintaining a shallow quantum circuit suitable for noisy intermediate-scale quantum (NISQ)-era constraints. Experimental evaluation on six MedMNIST benchmark datasets demonstrated competitive performance across both binary and multi-class classification tasks. HQCNN achieved 98.88% accuracy on the binary subset of PathMNIST (classes 0 vs. 1), 97.61% accuracy on the multi-class OrganAMNIST dataset, and 86.29% accuracy on BreastMNIST. Comparative experiments and statistical analyses demonstrated consistent improvements over the controlled BHQNN baseline, while component-wise ablation studies showed that the QAF module, superposition and entanglement mechanisms, and expressive parameterized rotations contributed to classification performance in a complementary and dataset-dependent manner. Moreover, HQCNN reduced the number of trainable quantum parameters by approximately 55.6% compared with the baseline hybrid quantum neural network (BHQNN). Noise-aware simulations further showed that the model retained relatively stable predictive performance under moderate depolarizing noise, supporting further evaluation under near-term quantum computing conditions. Overall, the results demonstrate that HQCNN provides a parameter-efficient hybrid quantum-classical framework for medical image classification and offers a promising foundation for further investigation of quantum-enhanced medical image analysis.
Robust medical image classification under input corruption and bag-level annotation remains a critical challenge in clinical AI applications. We propose QAPNet, a Quantum- Attentive Patchwise Network that integrates quantum neural encoding, additive attention-based instance reweighting, and prototype-contrastive regularization for reliable diagnosis from degraded inputs. Our framework uses a sliding-window strategy to divide each MRI medical Image into overlapping patches, where each is encoded via an 8-qubit quantum circuit using RY -based noise-sensitive layers for yielding expressive low-dimensional representations without relying on classical CNNs. A lightweight additive attention mechanism computes instance-wise importance weights that enable interpretable and noise-aware bag-level aggregation. To enhance robustness, we apply a contrastive loss that aligns clean and noisy embeddings and enforce prototype-guided clustering via class-wise centroids. We evaluate QAPNet across seven benchmark medical imaging datasets under three levels of additive Gaussian noise (σ ∈ {5%, 10%, 30%}). QAPNet consistently outperforms eight strong baselines and achieves up to +20.8% higher accuracy in OASIS (with 30% noise), +17.7% in PathMNIST, and maintains stable performance (< 4% degradation) in all settings. Ablation studies confirm the critical role of quantum encoding, attention-based aggregation, and prototype contrastive learning. These results suggest that QAPNet offers a scalable and interpretable architecture for noisy medical imaging tasks in the real world to bridge the quantum representation learning with robust clinical prediction.
Quantum image processing (QIP) is still a promising area as the noise of qubit, decoherence and limitations of encoding make it difficult for the current quantum hardware.This paper presents a combined Intelligent Quantum Image Denoising Framework, which uses FRQI and deep learning-based restoration, to enhance the quality and trustworthiness of images in quantum-classical hybrid areas.The short overview of related studies points out the classical filters and shallow neural models denoising methods, which are not able to generalize to quantum noise.To fill this void, a Convolutional Neural Network (CNN) is trained on the FRQI-encoded images simulated with Gaussian, saltand-pepper, and depolarizing noise.The approach uses confidence-based thresholding for restoration of adaptive pixels and measures the results with PSNR, SSIM, and MSE metrics.The experimental outcomes exhibit a considerable rise in the signal-to-noise ratio and the integrity of the structure when compared to the standard denoising techniques.The research indicates that the suggested combination of different methods has the capability to get rid of quantum noise yet at the same time allow the main image features to be seen.The steps for the next research will be applying the model to actual quantum simulators like Qiskit, creating better qubit usage, and widening the application of the framework to high-resolution images for quantum vision in real-world scenarios.
Quantum imaging systems produce images with distinctive noise patterns that conventional denoising algorithms cannot effectively process. We present an innovative neural network architecture that merges quantum physics principles with deep learning to address this challenge. Our hybrid approach adapts standard image processing techniques to handle quantum-specific noise while preserving critical image features. Experimental validation demonstrates a consistent 12.6% improvement in output quality compared to existing methods, with efficient performance on standard computing hardware. Additionally, the model exhibits strong generalization capabilities, achieving robust performance across varying noise levels. This advancement represents an important step toward practical quantum imaging applications in fields ranging from medical diagnostics to secure communications.
This paper examines a hybrid classical-quantum workflow for speckle-oriented image denoising, in which denoising is formulated as a discrete filter-parameter selection problem over image batches. Candidate convolution kernels are evaluated using batch-averaged peak signal-to-noise ratio (PSNR) and structural similarity (SSIM), and the best-performing kernel is deployed within a variance-guided adaptive filtering stage. The selected kernel configuration is then encoded into a quantum register and marked using a Grover oracle, followed by amplitude amplification to increase the likelihood of retrieving the target configuration upon measurement. In this formulation, the quantum circuit serves to demonstrate parameter encoding, oraclebased marking, and probabilistic retrieval consistent with the classically selected optimum, rather than replacing classical fitness evaluation. Experiments conducted on speckle-contaminated image batches from multiple imaging modalities show that the optimized kernel provides stable denoising performance, while the quantum subroutine reproduces the target configuration with high probability under practical circuit-depth constraints.
Introduction: Medical image classification is a crucial task in cancer diagnosis, relying on the accurate analysis of high-dimensional imaging data. While Convolutional Neural Networks (CNNs) have shown great success in this domain, their performance is often limited by the shallow feature expressiveness and overfitting, particularly in small or heterogeneous datasets. Methods: Quantum machine learning offers new opportunities through high-dimensional representations and nonlinear transformations. In this work, we propose a Quantum-Enhanced Sandwich Convolutional Neural Network (QSCNN), a layered hybrid architecture that integrates quantum and classical modules. The model employs a quanvolutional layer for localized quantum feature extraction, followed by conventional convolution and pooling for hierarchical representation learning, and a variational quantum classifier for nonlinear decision-making. Results: QSCNN achieved higher accuracy and training stability than classical CNNs and QCCNN baselines across three medical imaging tasks.: brain tumor MRI, skin cancer dermoscopy, and lung cancer CT. Circuit depth analysis revealed a trade-off between expressiveness and robustness, and additional experiments with depolarizing noise confirmed the model’s resilience under realistic quantum error conditions. Discussion: This suggests that circuit design choices influence hybrid model behavior and generalization, supporting the feasibility of quantum-enhanced methods for small-sample medical imaging. However, the current evaluation is limited to relatively small benchmark datasets, and broader validation on large-scale data will be essential to confirm clinical applicability. Conclusion: In summary, QSCNN presents a feasible hybrid framework for enhancing medical image classification with quantum features. While preliminary, our results suggest potential advantages in accuracy and stability under NISQ conditions.