Introduction: Brain stroke is still one of the leading causes of death and long-term disability in the world. Early and correct diagnosis is therefore important for patient outcome. Although Convolution Neural Network (CNN), classical machine learning models, have achieved great progress in medical image classification, they have to face the performance saturation problem when dealing with high-dimensional and complex data such as medical images. To tackle these limitations, we propose QBrainNet, a quantum enhanced model, which is to enhance brain stroke prediction from medical imaging datasets. Methods: The model consists of Quantum Neural Networks (QNNs) applied as learning complex patterns in terms of medical images and Variational Quantum Circuits (VQCs) that will be used to optimize the classification. The feature extraction featured in the QNNs utilises quantum properties of superposition and entanglement to extract non-linear high-dimensional patterns in images related to stroke that may not be captured using classical limits. The VQCs, in turn, are applied to optimize the model performance, further allocating the boundaries of the decision and enhancing the model performance in terms of accuracy by optimizing the quantum gates and operators used during the work. QBrainNet utilizes the combination of such quantum properties as entanglement and superposition to represent more complicated non-linear patterns in stroke-specific images in a better manner than a classical application does. Results: This paper proposes a hybrid classical-quantum scheme: preprocessing classically, and learning quantum-enhanced. Quantum gates and operators are used when performing the quantum phase to optimize decision boundaries, achieving vastly enhanced prediction accuracy and efficiency performance. Experimental results indicate that QBrainNet has a better accuracy (96%) and AUC-PR (0.97) than the classical models like CNN, SVM, and Random Forest, proving the superior performance of QBrainNet in stroke detection. Discussion: The inference time is shorter, so the model can be used as a real-time clinical application. This article points to the possibilities quantum computing can have in revolutionizing medical diagnostics, especially stroke prediction.
Introduction: Brain stroke is still one of the leading causes of death and long-term disability in the world. Early and correct diagnosis is therefore important for patient outcome. Although Convolution Neural Network (CNN), classical machine learning models, have achieved great progress in medical image classification, they have to face the performance saturation problem when dealing with high-dimensional and complex data such as medical images. To tackle these limitations, we propose QBrainNet, a quantum enhanced model, which is to enhance brain stroke prediction from medical imaging datasets. Methods: The model consists of Quantum Neural Networks (QNNs) applied as learning complex patterns in terms of medical images and Variational Quantum Circuits (VQCs) that will be used to optimize the classification. The feature extraction featured in the QNNs utilises quantum properties of superposition and entanglement to extract non-linear high-dimensional patterns in images related to stroke that may not be captured using classical limits. The VQCs, in turn, are applied to optimize the model performance, further allocating the boundaries of the decision and enhancing the model performance in terms of accuracy by optimizing the quantum gates and operators used during the work. QBrainNet utilizes the combination of such quantum properties as entanglement and superposition to represent more complicated non-linear patterns in stroke-specific images in a better manner than a classical application does. Results: This paper proposes a hybrid classical-quantum scheme: preprocessing classically, and learning quantum-enhanced. Quantum gates and operators are used when performing the quantum phase to optimize decision boundaries, achieving vastly enhanced prediction accuracy and efficiency performance. Experimental results indicate that QBrainNet has a better accuracy (96%) and AUC-PR (0.97) than the classical models like CNN, SVM, and Random Forest, proving the superior performance of QBrainNet in stroke detection. Discussion: The inference time is shorter, so the model can be used as a real-time clinical application. This article points to the possibilities quantum computing can have in revolutionizing medical diagnostics, especially stroke prediction.
Medical image classification is considered as very important field of diagnosis and treatment of neurological disorders, which include stroke, tumors, and hemorrhages, it can used to facilitate timely medical intervention. A hybrid quantum-classical convolutional neural network (QCNN) is presented by combining quantum information processing with classical deep learning techniques for improved feature extraction and classification accuracy. The model combines convolutional neural network (CNN), which handles initial feature extraction with a PennyLane and TensorFlow based layer that employs quantum entanglement and superposition principles to enhance classification performance. The model is trained and tested over a computed tomography (CT) scan image dataset which has four classes (normal, stroke, tumor, and hemorrhage). Various learning rates are tested and employed a hybrid backpropagation method to improve efficiency. Confusion matrices, region of conversion receiver operating characteristic (ROC) curves, and plots for the training convergence that all pointed to promising classification accuracy were derived with a detailed analysis accordingly. Overall, the results indicate that combining quantum computing with deep learning architectures could improve classification performance at a lowest computational cost. The results suggest the viability of hybrid quantum-classical models for medical imaging applications and indicate that quantum computing is a promising direction for enhancing diagnostic accuracy in the area of radiology.
Brain tumor detection is a vital and challenging task in medical diagnosis, requiring highly precise and efficient methods to improve patient treatment outcomes. Traditional approaches rely on classical machine learning and deep learning techniques, which, while effective to a degree, often struggle with capturing the complex and high-dimensional patterns present in medical imaging data. A significant limitation of these conventional methods is their inability to process and analyze intricate features efficiently, which can impact diagnostic accuracy. To address these challenges, this work proposes a novel Quantum Convolutional Neural Network (QCNN) framework that combines quantum computing principles, via quantum circuits, with classical machine learning architectures implemented through tools like PyTorch and PennyLane. This hybrid approach enhances feature extraction and classification capabilities, leading to superior performance. Experimental results demonstrate that the QCNN model achieves an improved accuracy of 90 percent on the test dataset, surpassing traditional methods and showing promise for advancing brain tumor detection technology. This improvement is supported by consistent decreases in training and validation loss during model training, highlighting the robustness and efficacy of the quantum-enhanced model.
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.
Quantum-Enhanced Machine Learning (QEML) is a new computational paradigm that introduces the concept of quantum mechanics with the latest artificial intelligence algorithms to enhance the performance of the diagnostic process in the field of medical imaging. The early identification of diseases is crucial in mortality reduction and improving patient outcome, especially in capacities dealing with such diseases as cancer, neurological diseases, heart diseases, and infections of the lungs. Although conventional deep learning architectures, in particular, Convolutional Neural Networks (CNNs) have been highly successful in processing complex medical images such as MRI scans, CT images, X-rays, and histopathological data, they can be computationally expensive, need large data sets, and suffer in optimization of high dimensional feature spaces. The limitations that QEML tries to overcome are based on the quantum mechanical principles of superposition, entanglement and interference to facilitate better data representation and processing. The suggested structure is based on the classical deep learning architectures and variational quantum circuits together, to create a hybrid quantum-classical system, which improves features extraction and classification levels. Under this method, the classical image characteristics are coded into quantum states and processed in high-dimensional Hilbert spaces, which can enhance the separability of subtle disease patterns, which would have otherwise been challenging to obtain with conventional techniques. The objective of this hybrid system is to enhance the critical diagnostic measures like sensitivity and specificity without affecting the interpretability of such measures to be used in the clinical context. The proposed framework is also assessed in terms of its feasibility, its computational performance and scalability to the capabilities of present-day Noisy Intermediate-Scale Quantum (NISQ) devices. The model effectiveness is evaluated using objective evaluation metrics, which, in a way, guarantee the feasibility of the model application within a real-life healthcare context. On balance, quantum computing and medical image analysis integration is an exciting field of work of the future intelligent healthcare systems and has a potential of providing more efficient, accurate, and scalable diagnostic solutions.
This study presents a comparative analysis of traditional deep learning and a novel quantum- assisted neural network for brain stroke identification on brain CT images. We compared traditional convolutional neural network (CNN) architectures such as ResNet18, VGG16, MobileNetV2 and In- ceptionV3. We found a hybrid quantum-classical approach which incorporated ResNet and MobileNet feature extraction tools and combined the extracted features in a quantum circuit for processing. The accuracy obtained for the quantum-assisted model on our validation dataset was 99.15%, which is a significant improvement from the traditional models: ResNet18 98.30%, VGG16 98.30%, MobileNetV2 97.02%, and InceptionV3 97.87%. These results indicate that there are likely advantages in enhanced medical image analysis by incorporating quantum techniques that may be useful in high stakes diagnos- tic like stroke detection. The hybrid architecture reflected some unique dynamic model assignment of importance for prediction assignment, with ResNet features contributing about 51.17% and MobileNet features contributing 48.83% to the overall predictions.
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.