Abstract In this paper, we propose a novel approach leveraging Quantum Generative Adversarial Networks (QGANs) for data augmentation in cancer image classification tasks. Cancer diagnosis from medical images plays a crucial role in early detection and treatment planning. However, imbalance nature of medical image data pose significant challenges for training accurate classification models. Our proposed methodology utilizes QGANs to generate synthetic medical images that closely mimic the characteristics of real cancer images classes, effectively addressing the data imbalance issue. Our QGAN architecture, which combines quantum circuits for the generator with a classical neural networks for the discriminator, was trained to synthesize medical images for augmentation. The performance of the convolutional neural networks (CNN) classifier was evaluated using accuracy and the area under the receiver operating characteristics curve (AUC-ROC) as metrics. Results show that augmented datasets significantly improve the classif ication performance. For instance, the AUC-ROC for the Basal Cell Carcinoma (BCC) class increased by 0.031, and the Dermatofibroma (DF) class saw a reduction in validation loss by 0.06. Moreover, segmentation-based QGAN training demonstrated considerable advantages in computational efficiency. The training time was reduced to just one hour, and inference times were less than a second per image.
Abstract In this paper, we propose a novel approach leveraging Quantum Generative Adversarial Networks (QGANs) for data augmentation in cancer image classification tasks. Cancer diagnosis from medical images plays a crucial role in early detection and treatment planning. However, imbalance nature of medical image data pose significant challenges for training accurate classification models. Our proposed methodology utilizes QGANs to generate synthetic medical images that closely mimic the characteristics of real cancer images classes, effectively addressing the data imbalance issue. Our QGAN architecture, which combines quantum circuits for the generator with a classical neural networks for the discriminator, was trained to synthesize medical images for augmentation. The performance of the convolutional neural networks (CNN) classifier was evaluated using accuracy and the area under the receiver operating characteristics curve (AUC-ROC) as metrics. Results show that augmented datasets significantly improve the classif ication performance. For instance, the AUC-ROC for the Basal Cell Carcinoma (BCC) class increased by 0.031, and the Dermatofibroma (DF) class saw a reduction in validation loss by 0.06. Moreover, segmentation-based QGAN training demonstrated considerable advantages in computational efficiency. The training time was reduced to just one hour, and inference times were less than a second per image.
Cancer detection using AI methods and image databases faces several challenges, including limited and imbalanced image datasets, long algorithm training times, and high-complexity feature generation. In particular, many cancer image datasets contain very few malignant examples compared to benign ones, limiting the amount of minority-class data available for training. Although classical AI and especially deep learning methods have had some success in providing cancer diagnostics, Quantum Machine Learning (QML) methods may offer advantages through high-dimensional feature representations, generative data augmentation, and the exploitation of feature correlations through quantum properties such as entanglement. In this paper, we propose using a generative QML method on the compressed breast ultrasound image dataset (BUSI) to generate synthetic images and hence augmented datasets for classifier training. In particular, we train a convolutional neural network with real, synthetic, and a combination of real and synthetic images. We test the algorithm on real images and provide comparative results for all three training cases. Our results show that the classifier trained entirely on synthetic images achieved an AUC-ROC of 0.802 (80.2%), compared with 0.760 (76.0%) for the real-data baseline.
This paper investigates the application of Quantum Generative Adversarial Networks (QGANs) to data augmentation for machine learning. Traditional Generative Adversarial Networks (GANs) rely on classical neural networks, while this research explores leveraging quantum circuits as generators within a QGAN architecture. The core idea is to train a quantum generator to produce synthetic data samples, supplementing existing datasets with novel instances. The discriminator remains a classical neural network, tasked with distinguishing between real and generated data. We present a theoretical framework for this approach, outlining the mathematical formulation and key components. Specifically, we introduce the notation (x, s_1, s_2, ..., s_K) to represent data samples, generator states, and relevant parameters. The objective function for the QGAN is defined as minimizing the Jensen-Shannon divergence between the real and generated data distributions. The effectiveness of this approach is discussed, highlighting potential advantages and challenges. This work contributes to the emerging field of quantum machine learning and offers a novel strategy for improving the performance of machine learning models, particularly in scenarios with limited training data.
This study investigates the application of a Advanced Intelligent Quantum Convolutional Neural Network (AI-QCNN) for classifying skin cancer images using the HAM10000 dataset. By integrating quantum computing with traditional CNN techniques, the model aims to enhance both classification accuracy and processing efficiency. During preprocessing, images undergo normalization, resizing, and augmentation to improve data quality and diversity, which helps the model to generalize more effectively and mitigate overfitting. The AIQCNN architecture combines classical convolutional layers for initial feature extraction with quantum circuits that further refine these features through quantum entanglement, resulting in improved classification outcomes. The AIQCNN achieves a training accuracy of 96% and a validation accuracy of 84%, demonstrating faster convergence and greater stability compared to conventional CNNs. These results highlight the potential of quantum computing to advance medical image classification and suggest opportunities for optimizing down sampling techniques and applying the model to various other medical imaging tasks.
Breast cancer is one of the most prevalent and life-threatening diseases affecting women worldwide, and early detection plays a crucial role in improving survival rates and treatment outcomes.Conventional diagnostic methods such as mammography, ultrasound imaging, and histopathological analysis rely heavily on expert interpretation and classical machine learning techniques.Although classical deep learning models, especially Convolutional Neural Networks (CNNs), have demonstrated impressive performance in medical image analysis, they face inherent limitations related to computational complexity, scalability, and the curse of dimensionality when handling large, high-resolution datasets.With the rapid advancement of quantum computing, Quantum Machine Learning (QML) has emerged as a promising paradigm capable of addressing these challenges by leveraging quantum phenomena such as superposition and entanglement.This research explores the application of Quantum Convolutional Neural Networks (Quantum CNNs or QCNNs) for breast cancer detection, aiming to enhance classification accuracy while reducing computational overhead.The proposed approach integrates quantum circuits with classical preprocessing techniques to extract discriminative features from breast cancer datasets, such as mammographic or histopathological images.By encoding image features into quantum states and performing convolution operations through parameterized quantum circuits, the Quantum CNN model is designed to learn complex patterns more efficiently than classical counterparts.The study evaluates the effectiveness of Quantum CNNs in distinguishing between benign and malignant breast tumors and compares its performance with traditional CNN models.The findings highlight the potential of Quantum CNNs to revolutionize medical image analysis by offering faster convergence, improved generalization, and robustness against noise.This research contributes to the growing field of quantumassisted healthcare systems and demonstrates how quantum computing can be harnessed to address critical challenges in breast cancer diagnosis.
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.
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.