Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Physica Scripta

Effective quantum convolutional neural network on symmetric complex-phase modulated RBS gate in pneumonia medical imaging diagnosis

Yan Liang, Wei Ding, Hong Xu, Haopeng Wang, Guoliang Shentu, Tianhui Qiu, Hongyang Ma

Published May 2, 2026• 1 citation
Publisher

Abstract

Abstract Pneumonia is a prevalent respiratory disease for which medical imaging plays an important role in diagnosis. With the growing integration of quantum computing and deep learning, Quantum Convolutional Neural Networks (QCNNs) have emerged as a promising framework for medical image analysis by exploiting quantum parallelism. However, their further development is constrained by the limitations of core quantum gates. To address key shortcomings in conventional QCNNs, particularly the error coupling caused by the asymmetric two-angle rotation in the Reconfigurable Beam Splitter (RBS) gate and the confinement of quantum states to real-valued space, this study proposes an optimized RBS gate with symmetric complex-phase modulation. Specifically, the proposed design replaces the asymmetric two-angle rotation with a symmetric single-angle rotation to alleviate rotation-mismatch-induced error coupling, and introduces an R z gate to enable complex-phase control. The optimized gate is then integrated into a baseline QCNN for pneumonia image diagnosis. Simulations on MNIST, Fashion-MNIST, and PneumoniaMNIST show that the proposed model achieves classification accuracies of 95.4%, 87.8%, and 82.1%, respectively, and outperforms the compared QCNN variants under the current setting. These findings suggest that gate-level optimization of the RBS gate may serve as an effective design strategy for enhancing QCNN-based medical image classification under controlled simulation settings.

DOI
10.1088/1402-4896/ae67b8

Related papers

Quanvolutional Neural Networks for Pneumonia Detection: An Efficient Quantum-Assisted Feature Extraction Paradigm

Gazi Tanbhir, Md. Farhan Shahriyar, Abdullah Md Raihan Chy

Pneumonia poses a significant global health challenge, demanding accurate and timely diagnosis. While deep learning, particularly Convolutional Neural Networks (CNNs), has shown promise in medical image analysis for pneumonia detection, CNNs often suffer from high computational costs, limitations in feature representation, and challenges in generalizing from smaller datasets. To address these limitations, we explore the application of Quanvolutional Neural Networks (QNNs), leveraging quantum computing for enhanced feature extraction. This paper introduces a novel hybrid quantum-classical model for pneumonia detection using the PneumoniaMNIST dataset. Our approach utilizes a quanvolutional layer with a parameterized quantum circuit (PQC) to process 2×2 image patches, employing rotational Y-gates for data encoding and entangling layers to generate non-classical feature representations. These quantum-extracted features are then fed into a classical neural network for classification. Experimental results demonstrate that the proposed QNN achieves a higher validation accuracy of 83.33% compared to a comparable classical CNN which achieves 73.33%. This enhanced convergence and sample efficiency highlight the potential of QNNs for medical image analysis, particularly in scenarios with limited labeled data. This research lays the foundation for integrating quantum computing into deep learning-driven medical diagnostic systems, offering a computationally efficient alternative to traditional approaches.

arXiv:2510.23660•International Conference on Next-Generation Computing, IoT and Machine Learning (NCIM)•Jun 27, 2025•6 citations

Automatic Classification of Medical Image Modality Using Quantum Convolutional Neural Network

Baraa Tantawi, Hamza Kamel Ahmed, Malak Magdy +1 more

Medical imaging plays an important role in the early diagnosis of various diseases. In machine learning algorithms, texture, color, and shape features are primarily used. This may result in a system with inadequate model generalization capabilities. The development of an end-to-end model that can classify unprocessed medical images is now possible by recent advances in deep learning. The heavy demands of deep learning on memory and computer resources result in further advancement in the processing of complex data. As a result, quantum computing offers potential solutions that take advantage of quantum mechanics concepts like entanglement, superposition, and interference. This chapter introduces a new medical image classification model based on the quantum convolutional neural network (QCNN). Three benchmark datasets are used to evaluate the performance of QCNN; PneumoniaMNIST, VesselMNIST3D, and brain tumor MRI images. The experimental results showed that the proposed model based on the proposed QCNN is very promising.

Advances in healthcare information systems and administration book series•Jun 30, 2023•1 citation

Hybrid Quantum-Classical Neural Networks for Medical Image Diagnosis

Mayank Genwa

Medical image classification is a fundamental aspect of contemporary diagnosis, but the conventional Deep Learning (DL) model is increasingly encountering difficulties in terms of computational complexity, data availability, and interpretability. With the massive growth of medical data, the "black box" phenomenon of Convolutional Neural Networks (CNNs) hinders medical validation. This paper presents a novel Hybrid Quantum-Classical Neural Network (HQCNN) approach for multimodal diagnosis, specifically addressing Brain Tumors (MRI), Pneumonia (Chest X-Ray), and Retinal Diseases (OCT). The proposed model combines a pre-trained ResNet-18 backbone for high-level feature representation with a 4-qubit Variational Quantum Circuit (VQC) to harness the power of quantum entanglement for improved classification in high-dimensional Hilbert spaces. More-over, we close the interpretability gap by incorporating Gradient-weighted Class Activation Mapping (Grad-CAM). Experimental results show that the hybrid model performs exceptionally well, specifically with 98.76% accuracy on Retina OCT, 90.71% accuracy on Chest X-Rays and 95.27% on Brain MRI, with a substantial decrease in trainable parameters in the decision head compared to classical models.

International Conference on Cognitive Computing and Networking Systems (ICC-CNS)•Jun 1, 2026•0 citations

Hybrid Quantum-Classical Convolutional Neural Networks with Gradient Variance-Controlled Optimization for Medical Images Classification

Mohammed Yousif, Mohammad Amjad Kamal

Medical image classification poses significant challenges due to the complex and nonlinear patterns present in imaging data. Traditional deep learning models often struggle to capture the intricate relationships inherent in such data, especially under conditions of noise and training instability. To address these limitations, this study proposes a novel Hybrid Quantum Convolutional Neural Network (HQCNN) integrated with a Gradient Variance-Controlled Optimizer (GVCO). The aim is to enhance diagnostic accuracy by combining classical convolutional layers for shallow feature extraction with a parameterized quantum circuit (PQC) that enables richer and more discriminative feature representations. The GVCO dynamically adjusts learning rates and momentum coefficients based on real-time gradient variance, improving convergence speed and generalization performance. The proposed HQCNN-GVCO model was evaluated on two public brain imaging datasets: the Kaggle Brain MRI dataset and the REMBRANDT dataset. It achieved classification accuracies of 99.2% and 98.2%, respectively. Compared to baseline HQCNN and other state-of-the-art models, HQCNN-GVCO demonstrated superior and more consistent performance. These findings highlight the potential of integrating quantum computing techniques with adaptive optimization strategies to advance reliable and accurate medical image analysis.

Journal of Artificial Intelligence in Medical Applications•Sep 16, 2025•1 citation

Q-BacilliNet: A Hybrid Quantum Convolutional Neural Network for Automated Tuberculosis Bacilli Detection

Nagaraju Y, Mohammed Imaad Sharieff, Mayank V K +2 more

Tuberculosis (TB) remains a leading infectious disease burden in resource-limited healthcare settings, where timely and accurate diagnosis from Ziehl–Neelsen sputum microscopy is critical yet constrained by manual inspection challenges. Existing deep learning approaches, while effective, suffer from high computational demands and limited generalization under class-imbalanced medical datasets. In this paper, we propose Q-BacilliNet, a hybrid Quantum Convolutional Neural Network (QCNN) that integrates classical CNN-based spatial feature extraction with a 5-qubit variational quantum circuit for feature enhancement, followed by fully connected classification layers. The quantum layer encodes compressed CNN features into quantum states via parameterized RY rotation gates, applies entanglement through circular CNOT patterns, and extracts Pauli-Z expectation values as quantum-enhanced representations. To address class imbalance inherent in TB microscopy datasets, the framework employs focal loss, diversity regularization, and balanced batch sampling. The proposed NISQ-compatible architecture is evaluated on the Ziehl–Neelsen Sputum Microscopy Image Database (ZNSM-iDB) and benchmarked against MobileNet and EfficientNet baselines. The 5-qubit Q-BacilliNet achieves 88.0% overall accuracy, 83.1% TB detection accuracy, and 91.6% balanced accuracy, outperforming all baselines. Ablation experiments comparing 4-qubit and 5-qubit configurations validate the benefit of increased quantum expressibility. These results demonstrate the viability of hybrid quantum–classical architectures for automated TB diagnosis in low-resource clinical environments.

International Conference on Intelligent Technologies (CONIT)•Jun 1, 2026•0 citations

A Review on Image Classification of Medical Images for Tumor Detection Using Quantum Convolutional Neural Networks

Priyanka Killedar, Vibha Tiwari

Quantum Convolutional Neural Networks (QCNNs) have became a promising field for quantum advantage in feature extraction and classification of different datasets like Handwritten datasets, Fashion datasets, particularly useful for complex datasets like medical images. Medical image classification has emerged as a critical component in modern healthcare diagnostics, particularly for tumor detection and cancer diagnosis. Normally, Classical CNN is used for image classification but they may face certain challenges in achieving further improvements in accuracy, computational efficiency, processing time. Also, it becomes difficult to train for high-dimensional medical datasets and extracting complex feature representations. Therefore, In this proposal, Image classification through a Scalable and Resource efficient QCNN is proposed which integrates classical CNN with quantum computing. This review examines the current state of medical image classification for tumor detection, analyzes the transition from classical CNNs to quantum-based architectures, and explores the potential of QCNNs in revolutionizing medical diagnostics. This research includes QCNN architecture which will include selective feature encoding using encoding technique such as Z Feature map and quantum circuits for convolutional layers and pooling layers to handle multi-modal medical data (e.g., MRI and CT scans). We aim to achieve superior accuracy with reduced parameters over classical CNNs. This work advances quantum ML toward practical deployment in resource-limited healthcare settings.

International Journal for Research in Applied Science and Engineering Technology•Dec 19, 2025•0 citations
Abstract Pneumonia is a prevalent respiratory disease for which medical imaging plays an important role in diagnosis. With the growing integration of quantum computing and deep learning, Quantum Convolutional Neural Networks (QCNNs) have emerged as a promising framework for medical image analysis by exploiting quantum parallelism. However, their further development is constrained by the limitations of core quantum gates. To address key shortcomings in conventional QCNNs, particularly the error coupling caused by the asymmetric two-angle rotation in the Reconfigurable Beam Splitter (RBS) gate and the confinement of quantum states to real-valued space, this study proposes an optimized RBS gate with symmetric complex-phase modulation. Specifically, the proposed design replaces the asymmetric two-angle rotation with a symmetric single-angle rotation to alleviate rotation-mismatch-induced error coupling, and introduces an R z gate to enable complex-phase control. The optimized gate is then integrated into a baseline QCNN for pneumonia image diagnosis. Simulations on MNIST, Fashion-MNIST, and PneumoniaMNIST show that the proposed model achieves classification accuracies of 95.4%, 87.8%, and 82.1%, respectively, and outperforms the compared QCNN variants under the current setting. These findings suggest that gate-level optimization of the RBS gate may serve as an effective design strategy for enhancing QCNN-based medical image classification under controlled simulation settings.
Pneumonia poses a significant global health challenge, demanding accurate and timely diagnosis. While deep learning, particularly Convolutional Neural Networks (CNNs), has shown promise in medical image analysis for pneumonia detection, CNNs often suffer from high computational costs, limitations in feature representation, and challenges in generalizing from smaller datasets. To address these limitations, we explore the application of Quanvolutional Neural Networks (QNNs), leveraging quantum computing for enhanced feature extraction. This paper introduces a novel hybrid quantum-classical model for pneumonia detection using the PneumoniaMNIST dataset. Our approach utilizes a quanvolutional layer with a parameterized quantum circuit (PQC) to process 2×2 image patches, employing rotational Y-gates for data encoding and entangling layers to generate non-classical feature representations. These quantum-extracted features are then fed into a classical neural network for classification. Experimental results demonstrate that the proposed QNN achieves a higher validation accuracy of 83.33% compared to a comparable classical CNN which achieves 73.33%. This enhanced convergence and sample efficiency highlight the potential of QNNs for medical image analysis, particularly in scenarios with limited labeled data. This research lays the foundation for integrating quantum computing into deep learning-driven medical diagnostic systems, offering a computationally efficient alternative to traditional approaches.
Medical imaging plays an important role in the early diagnosis of various diseases. In machine learning algorithms, texture, color, and shape features are primarily used. This may result in a system with inadequate model generalization capabilities. The development of an end-to-end model that can classify unprocessed medical images is now possible by recent advances in deep learning. The heavy demands of deep learning on memory and computer resources result in further advancement in the processing of complex data. As a result, quantum computing offers potential solutions that take advantage of quantum mechanics concepts like entanglement, superposition, and interference. This chapter introduces a new medical image classification model based on the quantum convolutional neural network (QCNN). Three benchmark datasets are used to evaluate the performance of QCNN; PneumoniaMNIST, VesselMNIST3D, and brain tumor MRI images. The experimental results showed that the proposed model based on the proposed QCNN is very promising.
Medical image classification is a fundamental aspect of contemporary diagnosis, but the conventional Deep Learning (DL) model is increasingly encountering difficulties in terms of computational complexity, data availability, and interpretability. With the massive growth of medical data, the "black box" phenomenon of Convolutional Neural Networks (CNNs) hinders medical validation. This paper presents a novel Hybrid Quantum-Classical Neural Network (HQCNN) approach for multimodal diagnosis, specifically addressing Brain Tumors (MRI), Pneumonia (Chest X-Ray), and Retinal Diseases (OCT). The proposed model combines a pre-trained ResNet-18 backbone for high-level feature representation with a 4-qubit Variational Quantum Circuit (VQC) to harness the power of quantum entanglement for improved classification in high-dimensional Hilbert spaces. More-over, we close the interpretability gap by incorporating Gradient-weighted Class Activation Mapping (Grad-CAM). Experimental results show that the hybrid model performs exceptionally well, specifically with 98.76% accuracy on Retina OCT, 90.71% accuracy on Chest X-Rays and 95.27% on Brain MRI, with a substantial decrease in trainable parameters in the decision head compared to classical models.
Medical image classification poses significant challenges due to the complex and nonlinear patterns present in imaging data. Traditional deep learning models often struggle to capture the intricate relationships inherent in such data, especially under conditions of noise and training instability. To address these limitations, this study proposes a novel Hybrid Quantum Convolutional Neural Network (HQCNN) integrated with a Gradient Variance-Controlled Optimizer (GVCO). The aim is to enhance diagnostic accuracy by combining classical convolutional layers for shallow feature extraction with a parameterized quantum circuit (PQC) that enables richer and more discriminative feature representations. The GVCO dynamically adjusts learning rates and momentum coefficients based on real-time gradient variance, improving convergence speed and generalization performance. The proposed HQCNN-GVCO model was evaluated on two public brain imaging datasets: the Kaggle Brain MRI dataset and the REMBRANDT dataset. It achieved classification accuracies of 99.2% and 98.2%, respectively. Compared to baseline HQCNN and other state-of-the-art models, HQCNN-GVCO demonstrated superior and more consistent performance. These findings highlight the potential of integrating quantum computing techniques with adaptive optimization strategies to advance reliable and accurate medical image analysis.
Tuberculosis (TB) remains a leading infectious disease burden in resource-limited healthcare settings, where timely and accurate diagnosis from Ziehl–Neelsen sputum microscopy is critical yet constrained by manual inspection challenges. Existing deep learning approaches, while effective, suffer from high computational demands and limited generalization under class-imbalanced medical datasets. In this paper, we propose Q-BacilliNet, a hybrid Quantum Convolutional Neural Network (QCNN) that integrates classical CNN-based spatial feature extraction with a 5-qubit variational quantum circuit for feature enhancement, followed by fully connected classification layers. The quantum layer encodes compressed CNN features into quantum states via parameterized RY rotation gates, applies entanglement through circular CNOT patterns, and extracts Pauli-Z expectation values as quantum-enhanced representations. To address class imbalance inherent in TB microscopy datasets, the framework employs focal loss, diversity regularization, and balanced batch sampling. The proposed NISQ-compatible architecture is evaluated on the Ziehl–Neelsen Sputum Microscopy Image Database (ZNSM-iDB) and benchmarked against MobileNet and EfficientNet baselines. The 5-qubit Q-BacilliNet achieves 88.0% overall accuracy, 83.1% TB detection accuracy, and 91.6% balanced accuracy, outperforming all baselines. Ablation experiments comparing 4-qubit and 5-qubit configurations validate the benefit of increased quantum expressibility. These results demonstrate the viability of hybrid quantum–classical architectures for automated TB diagnosis in low-resource clinical environments.
Quantum Convolutional Neural Networks (QCNNs) have became a promising field for quantum advantage in feature extraction and classification of different datasets like Handwritten datasets, Fashion datasets, particularly useful for complex datasets like medical images. Medical image classification has emerged as a critical component in modern healthcare diagnostics, particularly for tumor detection and cancer diagnosis. Normally, Classical CNN is used for image classification but they may face certain challenges in achieving further improvements in accuracy, computational efficiency, processing time. Also, it becomes difficult to train for high-dimensional medical datasets and extracting complex feature representations. Therefore, In this proposal, Image classification through a Scalable and Resource efficient QCNN is proposed which integrates classical CNN with quantum computing. This review examines the current state of medical image classification for tumor detection, analyzes the transition from classical CNNs to quantum-based architectures, and explores the potential of QCNNs in revolutionizing medical diagnostics. This research includes QCNN architecture which will include selective feature encoding using encoding technique such as Z Feature map and quantum circuits for convolutional layers and pooling layers to handle multi-modal medical data (e.g., MRI and CT scans). We aim to achieve superior accuracy with reduced parameters over classical CNNs. This work advances quantum ML toward practical deployment in resource-limited healthcare settings.