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

Diagnostics

Does the Quantum Layer Help? A Leakage-Aware Benchmark of Hybrid Quantum–Classical Models for Brain Tumor MRI Classification

Linh T. Duong, Pande Wikantyasa Maswita, Tony Ibnu Sumaryada, Sitti Yani, Dương Thanh Tài, Hiba Baha Omer, Nissren M. Tamam, Abdelmoneim Sulieman

Published Oct 24, 2026• 0 citations
Publisher

Abstract

Background/Objectives: Hybrid quantum–classical neural networks are reported to classify brain tumors on magnetic resonance imaging (MRI) accurately, but these reports rarely control for data leakage or test whether the quantum circuit contributes. We examined both issues. Methods: We audited version 2 of the Kaggle Brain Tumor MRI Dataset (7200 images) with mirror- and rotation-invariant duplicate detection and by matching every image to its source collections, which recovered patient identities for 4846 tumor images. A variational quantum circuit (VQC) was compared with size-matched, deeper and Fourier-surrogate classical controls using frozen features, end-to-end fine-tuning and training from scratch, all under patient-grouped five-fold cross-validation. Results: Of the 1600 test images, 1269 (79%) had a copy in the training partition, and 915 of the 917 figshare-derived test images shared a patient with it. On frozen features, a six-qubit VQC head was 6.0 percentage points (pp) less accurate than its classical control (95% confidence interval −6.8 to −5.2) and less accurate than a classical surrogate of its own function class. With fine-tuning, it stayed within about 1 pp of its control but was 0.7 pp less accurate (−1.2 to −0.1) after convergence-based training; a hybrid network trained from scratch was 8.3 pp less accurate than its control. Patient-grouped evaluation lowered accuracy by 3–8 pp relative to our earlier duplicate-grouped evaluation, and 76 glioma test images from another source remained poorly recognized. Conclusions: The quantum layer added cost but no accuracy, and leakage and provenance mattered more than the classifier head.

DOI
10.3390/diagnostics16213485

Related papers

Quantum Convolutional Neural Networks for Brain Tumor MRI Image Classification: Architecture, Implementation, and Performance Analysis

Priyanka Killedar, Vibha Tiwari

Magnetic Resonance Imaging (MRI) scans of the brain can serve as a clinically vital task computationally challenging. Regular Convolutional Neural Networks (CNNs) are highly accurate, but they require significant computational power, thus restricting their use in resource-limited clinical settings. This work presents a hybrid Quantum Convolutional Neural Network (QCNN) architecture that combines parameterized quantum circuits (PQCs) with classical CNN layers to boost the performance of the classification task and minimize the set of trainable parameters. The proposed architecture uses variational quantum circuits (VQCs) as quantum feature extractors in intermediate processing layers, and final classification (classical fully connected layers). The parameter-shift rule trains quantum circuit parameters end-to-end and allows calculating the gradient without finite-differentiating the circuits. Three publicly available benchmark datasets, including the Kaggle Brain Tumor MRI Dataset, the BraTS 2020 Challenge Dataset, and Brain MRI Segmentation Dataset, which comprise four pathological categories, glioma, meningioma, pituitary adenoma, and no-tumor, were experimented on. The proposed QCNN reached a classification accuracy of 97.84%, sensitivity of 97.21, specificity of 98.43 and F1-score of 97.52 on the Kaggle test partition, which is better than five classical CNN baselines, such as ResNet-50, VGG-16, EfficientNet-B4, DenseNet-121, and In Moreover, the QCNN showed a 74.6% reduction in trainable parameters over EfficientNet-B4, which indicates the practical efficiency of quantum-classical hybrid networks in medical image analysis.

IEEE International Conference on Advances in Computing Research on Science Engineering and Technology (ACROSET)•Sep 1, 2026•0 citations

Hybrid Quantum-Classical Model for Multi-Class Brain Tumor MRI Classification

Singampalli Hemanth Kumar, Hazarath Murali Krishna Prasad

The problem of classifying brain tumours using magnetic resonance imaging (MRI) is a fundamental one in the field of computeraided diagnosis, since although many commonly employed convolutional neural network (CNN) architectures provide high accuracy they contain a large number of parameters, which has led to a search for more compact network designs. In this paper a hybrid quantum–classical approach is proposed that combines a lightweight CNN feature extractor with a trainable variational quantum circuit (VQC) in order to carry out four-class brain tumour classification. The MRI images are resized to 128 by 128 pixels and then processed by a small three-block CNN that reduces each image to an 8-dimensional feature vector. This vector is put into an 8-qubit quantum register by means of angle embedding and then acted on by a variational circuit composed of strongly entangling layers, by measuring each qubit in the Pauli-Z basis. The quantum features thus obtained are then fed into classical dense layers that categorise the images as glioma, meningioma, pituitary tumour or no tumour. The framework was tested on the Brain Tumor MRI dataset (7,200 images) and compared with a classical CNN baseline that was trained under the same conditions. The hybrid model reached a test accuracy of 76.88%, as compared to 72.31% for the classical baseline an improvement of 4.57 percentage points together with corresponding increases in macro-averaged precision, recall and F1-score. Notably, the hybrid model needed only 72 additional trainable parameters when compared to the classical baseline. The fact that the quantum layer was assessed using a noiseless simulator rather than on actual quantum hardware means that these results should be seen as indicating the feasibility of using compact quantum feature transformations for medical image classification rather than as showing a quantum advantage.

•Jan 1, 2026•0 citations

Hybrid CNN–VQC for Brain Tumor MRI Classification

P Sree Manvitha, Ramdas Kapila

Classification of brain tumors from Magnetic Resonance Imaging (MRI) is a critical step in clinical diagnosis and treatment planning. While Convolutional Neural Networks (CNNs) achieve strong performance in medical imaging, they often suffer from class imbalance and limited generalization when trained on small datasets. This paper proposes a hybrid Convolutional Neural Network–Variational Quantum Classifier (CNN–VQC) framework for multiclass brain tumor classification. MRI scans are first preprocessed and normalized, after which an EfficientNet-B0 backbone extracts deep feature representations. Class imbalance is then addressed directly in the feature space by combining the Synthetic Minority Over-sampling Technique with Tomek-link cleaning (SMOTE-TOMEK), followed by Principal Component Analysis (PCA) for dimensionality reduction to enable quantum encoding. Finally, a variational quantum circuit coupled with a softmax layer performs the classification. The framework is evaluated on two benchmark datasets: the Kaggle Brain Tumor MRI dataset (Nickparvar, 7,023 images, four classes) and the Figshare Brain Tumor Dataset (Cheng et al., 3,064 images, three classes). Across five stratified runs, the proposed model attains accuracies of $94.8 \pm 0.4 \%$ and $93.1 \pm 0.5 \%$ respectively, with macro F1-scores of 0.941 and 0.927, outperforming EfficientNet-B0, ResNet-18, and standalone VQC baselines with statistically significant margins (paired t-test, $\mathrm{p}\lt 0.01$). Additional analyses of computational complexity, inference latency, and noisy quantum simulation demonstrate the practical feasibility of the proposed approach on near-term quantum hardware.

International Conference on Innovative Mechanisms for Industry Applications•Oct 1, 2026•0 citations

A Staged Ablation Study of Quantum Convolution and Variational Quantum Classification Placement in Hybrid CNNs for Brain Tumor MRI Classification

C. Ramachandran, Dr. V. Kathiresan

Abstract Quantum machine learning (QML) has recently been explored as a means of augmenting classical convolutional neural networks (CNNs) for medical image analysis, yet most existing hybrid architectures insert a single quantum component either at the input or the output of an otherwise classical pipeline, and few works empirically isolate which placement is actually responsible for performance change. This paper proposes QCQ-CNN, a Quantum-Classical-Quantum hybrid framework for brain tumor magnetic resonance imaging (MRI) classification in which a quantum convolutional (“quanvolutional”) layer performs early-stage feature extraction, a classical CNN backbone performs hierarchical representation learning, and a variational quantum circuit (VQC) performs final classification, so that quantum processing brackets the classical backbone at both ends. Using a real, held-out four-class brain tumor MRI dataset (glioma, meningioma, pituitary tumor, and no tumor), we implement and train four architectural variants under matched conditions — (A) a fully classical CNN baseline, (B) a quantum-convolution-plus-classical-CNN variant, (C) a classical-CNN-plus-quantum-head variant, and (D) the full proposed QCQ-CNN — and report actual experimental accuracy, precision, recall, and F1-score on a common test set, together with training curves, confusion matrices, and qualitative visualizations of the learned quantum feature maps. Our small-scale experiments show that quantum convolutional preprocessing alone improves macro-F1 over the classical baseline (0.204 to 0.382), while replacing the classical classification head with a variational quantum circuit degrades performance in both the quantum-head-only and the full QCQ-CNN configurations. We discuss this finding in the context of variational quantum circuit trainability at low qubit counts and limited data, and argue that our staged ablation methodology — rather than a single best-accuracy claim — constitutes the paper's principal novelty: a reusable protocol for empirically localizing where, in a CNN pipeline, quantum layers help versus hurt. We conclude with concrete design recommendations and directions for scaling QCQ-style frameworks. Keywords: Quantum Machine Learning, Hybrid Quantum-Classical Neural Networks, Quanvolutional Neural Network, Variational Quantum Circuit, Convolutional Neural Network, Brain Tumor Classification, Medical Image Processing

International Scientific Journal of Engineering and Management•Aug 20, 2026•0 citations

Hybrid quantum–classical learning for MRI-based brain tumour diagnosis

A. Harshavardhan, V. Chandra Shekhar Rao, Y. Madhavi Reddy +3 more

Accurate classification of glioma grades from magnetic resonance imaging (MRI) is essential for clinical decision-making in neuro-oncology. Although deep learning performance has been impressive with classical models, they struggle with high-dimensional medical imaging data and generalise poorly beyond their training data, especially in time- and resource-constrained settings. In light of the aforementioned challenges, we propose QuantumMedDx, a hybrid quantum–classical learning framework for classifying gliomas using MRI. The framework combines quantum feature encoding and variational quantum circuits with classical neural inference to improve diagnostic performance. The base model, QImageNet, uses amplitude-based quantum encoding for writing, entanglement-enabled parameterised quantum circuits (EPQCs) as feature extractors, and classical dense layers for classifying HGG and LGG from multimodal MRI slices. We demonstrate the effectiveness of the proposed approach on the BraTS 2021 benchmark dataset using a patient-aware 5-fold cross-validation protocol. Experimental results show that QuantumMedDx achieves accuracies of 94.12%, 93.30%, and 96.42%; F1-scores of 93.30% and 96.42%; and AUCs of 96.42% and 96.42%, respectively, outperforming conventional CNN, DNN, and SVM baselines. Ablation studies provide additional evidence of the performance improvements enabled by quantum Fourier transform and entanglement layers. Such results suggest that quantum–classical learning can efficiently improve feature extraction and discrimination in medical imaging, thus providing a modular and scalable route towards quantum–inspired clinical decision–support systems of the future.

Discover Computing•May 14, 2026•2 citations

Quantum Kernel-Driven Hybrid Neural Network for Brain Tumor Classification

Alok Kumar Srivastava, Shadab Hussain, Shiru Sharma +2 more

Multiclass brain tumor classification from magnetic resonance imaging (MRI) remains challenging due to subtle inter-class visual similarities, particularly for anatomically small and heterogeneous tumor types such as pituitary lesions. To address these challenges, this paper presents a hybrid classical-quantum learning framework that integrates a deep convolutional feature extractor from the ResNet family with a variational quantum neural network (QNN) head for enhanced feature representation. In the present work, classical deep features extracted by the ResNet model are encoded into a six-qubit quantum circuit, enabling richer feature interactions within a high-dimensional Hilbert space. In addition, an anatomically guided Kernel-based Center Cropping (KCC) strategy is introduced as a deterministic spatial prior to emphasize anatomically relevant central brain structures, particularly benefiting challenging pituitary tumor cases. The proposed Kernel-based Center Cropping–Hybrid Quantum Neural Network (KCC-HQNN) framework is trained and evaluated on a publicly available composite four-class brain MRI dataset curated from Kaggle sources, comprising glioma, meningioma, pituitary tumor, and no-tumor categories. Experimental results across multiple ResNet variants demonstrate that the hybrid quantum–classical models consistently outperform their purely classical counterparts, achieving classification accuracy of up to 99%, while macro-averaged precision, recall, and F1 scores consistently fall in the range of approximately 0.95–0.99 across different ResNet variants. Reliability and reproducibility are further validated through stratified five-fold cross-validation and repeated experiments using multiple random initialization seeds, yielding a mean classification accuracy of 99.00 ± 0.08%, macro F1-score of 0.986 ± 0.11, and weighted F1-score of 0.989 ± 0.12, while cross-dataset evaluation on independent IEEE DataPort MRI datasets demonstrates robust generalization under dataset shift. All experiments are conducted on publicly available retrospective datasets using a quantum simulator. These results indicate that combining quantum feature processing with anatomically informed spatial processing improves multiclass discrimination of brain tumors.

IEEE Access•Jan 1, 2026•3 citations
Background/Objectives: Hybrid quantum–classical neural networks are reported to classify brain tumors on magnetic resonance imaging (MRI) accurately, but these reports rarely control for data leakage or test whether the quantum circuit contributes. We examined both issues. Methods: We audited version 2 of the Kaggle Brain Tumor MRI Dataset (7200 images) with mirror- and rotation-invariant duplicate detection and by matching every image to its source collections, which recovered patient identities for 4846 tumor images. A variational quantum circuit (VQC) was compared with size-matched, deeper and Fourier-surrogate classical controls using frozen features, end-to-end fine-tuning and training from scratch, all under patient-grouped five-fold cross-validation. Results: Of the 1600 test images, 1269 (79%) had a copy in the training partition, and 915 of the 917 figshare-derived test images shared a patient with it. On frozen features, a six-qubit VQC head was 6.0 percentage points (pp) less accurate than its classical control (95% confidence interval −6.8 to −5.2) and less accurate than a classical surrogate of its own function class. With fine-tuning, it stayed within about 1 pp of its control but was 0.7 pp less accurate (−1.2 to −0.1) after convergence-based training; a hybrid network trained from scratch was 8.3 pp less accurate than its control. Patient-grouped evaluation lowered accuracy by 3–8 pp relative to our earlier duplicate-grouped evaluation, and 76 glioma test images from another source remained poorly recognized. Conclusions: The quantum layer added cost but no accuracy, and leakage and provenance mattered more than the classifier head.
Magnetic Resonance Imaging (MRI) scans of the brain can serve as a clinically vital task computationally challenging. Regular Convolutional Neural Networks (CNNs) are highly accurate, but they require significant computational power, thus restricting their use in resource-limited clinical settings. This work presents a hybrid Quantum Convolutional Neural Network (QCNN) architecture that combines parameterized quantum circuits (PQCs) with classical CNN layers to boost the performance of the classification task and minimize the set of trainable parameters. The proposed architecture uses variational quantum circuits (VQCs) as quantum feature extractors in intermediate processing layers, and final classification (classical fully connected layers). The parameter-shift rule trains quantum circuit parameters end-to-end and allows calculating the gradient without finite-differentiating the circuits. Three publicly available benchmark datasets, including the Kaggle Brain Tumor MRI Dataset, the BraTS 2020 Challenge Dataset, and Brain MRI Segmentation Dataset, which comprise four pathological categories, glioma, meningioma, pituitary adenoma, and no-tumor, were experimented on. The proposed QCNN reached a classification accuracy of 97.84%, sensitivity of 97.21, specificity of 98.43 and F1-score of 97.52 on the Kaggle test partition, which is better than five classical CNN baselines, such as ResNet-50, VGG-16, EfficientNet-B4, DenseNet-121, and In Moreover, the QCNN showed a 74.6% reduction in trainable parameters over EfficientNet-B4, which indicates the practical efficiency of quantum-classical hybrid networks in medical image analysis.
The problem of classifying brain tumours using magnetic resonance imaging (MRI) is a fundamental one in the field of computeraided diagnosis, since although many commonly employed convolutional neural network (CNN) architectures provide high accuracy they contain a large number of parameters, which has led to a search for more compact network designs. In this paper a hybrid quantum–classical approach is proposed that combines a lightweight CNN feature extractor with a trainable variational quantum circuit (VQC) in order to carry out four-class brain tumour classification. The MRI images are resized to 128 by 128 pixels and then processed by a small three-block CNN that reduces each image to an 8-dimensional feature vector. This vector is put into an 8-qubit quantum register by means of angle embedding and then acted on by a variational circuit composed of strongly entangling layers, by measuring each qubit in the Pauli-Z basis. The quantum features thus obtained are then fed into classical dense layers that categorise the images as glioma, meningioma, pituitary tumour or no tumour. The framework was tested on the Brain Tumor MRI dataset (7,200 images) and compared with a classical CNN baseline that was trained under the same conditions. The hybrid model reached a test accuracy of 76.88%, as compared to 72.31% for the classical baseline an improvement of 4.57 percentage points together with corresponding increases in macro-averaged precision, recall and F1-score. Notably, the hybrid model needed only 72 additional trainable parameters when compared to the classical baseline. The fact that the quantum layer was assessed using a noiseless simulator rather than on actual quantum hardware means that these results should be seen as indicating the feasibility of using compact quantum feature transformations for medical image classification rather than as showing a quantum advantage.
Classification of brain tumors from Magnetic Resonance Imaging (MRI) is a critical step in clinical diagnosis and treatment planning. While Convolutional Neural Networks (CNNs) achieve strong performance in medical imaging, they often suffer from class imbalance and limited generalization when trained on small datasets. This paper proposes a hybrid Convolutional Neural Network–Variational Quantum Classifier (CNN–VQC) framework for multiclass brain tumor classification. MRI scans are first preprocessed and normalized, after which an EfficientNet-B0 backbone extracts deep feature representations. Class imbalance is then addressed directly in the feature space by combining the Synthetic Minority Over-sampling Technique with Tomek-link cleaning (SMOTE-TOMEK), followed by Principal Component Analysis (PCA) for dimensionality reduction to enable quantum encoding. Finally, a variational quantum circuit coupled with a softmax layer performs the classification. The framework is evaluated on two benchmark datasets: the Kaggle Brain Tumor MRI dataset (Nickparvar, 7,023 images, four classes) and the Figshare Brain Tumor Dataset (Cheng et al., 3,064 images, three classes). Across five stratified runs, the proposed model attains accuracies of 94.8±0.4%94.8 \pm 0.4 \%94.8±0.4% and 93.1±0.5%93.1 \pm 0.5 \%93.1±0.5% respectively, with macro F1-scores of 0.941 and 0.927, outperforming EfficientNet-B0, ResNet-18, and standalone VQC baselines with statistically significant margins (paired t-test, p<0.01\mathrm{p}\lt 0.01p<0.01). Additional analyses of computational complexity, inference latency, and noisy quantum simulation demonstrate the practical feasibility of the proposed approach on near-term quantum hardware.
Abstract Quantum machine learning (QML) has recently been explored as a means of augmenting classical convolutional neural networks (CNNs) for medical image analysis, yet most existing hybrid architectures insert a single quantum component either at the input or the output of an otherwise classical pipeline, and few works empirically isolate which placement is actually responsible for performance change. This paper proposes QCQ-CNN, a Quantum-Classical-Quantum hybrid framework for brain tumor magnetic resonance imaging (MRI) classification in which a quantum convolutional (“quanvolutional”) layer performs early-stage feature extraction, a classical CNN backbone performs hierarchical representation learning, and a variational quantum circuit (VQC) performs final classification, so that quantum processing brackets the classical backbone at both ends. Using a real, held-out four-class brain tumor MRI dataset (glioma, meningioma, pituitary tumor, and no tumor), we implement and train four architectural variants under matched conditions — (A) a fully classical CNN baseline, (B) a quantum-convolution-plus-classical-CNN variant, (C) a classical-CNN-plus-quantum-head variant, and (D) the full proposed QCQ-CNN — and report actual experimental accuracy, precision, recall, and F1-score on a common test set, together with training curves, confusion matrices, and qualitative visualizations of the learned quantum feature maps. Our small-scale experiments show that quantum convolutional preprocessing alone improves macro-F1 over the classical baseline (0.204 to 0.382), while replacing the classical classification head with a variational quantum circuit degrades performance in both the quantum-head-only and the full QCQ-CNN configurations. We discuss this finding in the context of variational quantum circuit trainability at low qubit counts and limited data, and argue that our staged ablation methodology — rather than a single best-accuracy claim — constitutes the paper's principal novelty: a reusable protocol for empirically localizing where, in a CNN pipeline, quantum layers help versus hurt. We conclude with concrete design recommendations and directions for scaling QCQ-style frameworks. Keywords: Quantum Machine Learning, Hybrid Quantum-Classical Neural Networks, Quanvolutional Neural Network, Variational Quantum Circuit, Convolutional Neural Network, Brain Tumor Classification, Medical Image Processing
Accurate classification of glioma grades from magnetic resonance imaging (MRI) is essential for clinical decision-making in neuro-oncology. Although deep learning performance has been impressive with classical models, they struggle with high-dimensional medical imaging data and generalise poorly beyond their training data, especially in time- and resource-constrained settings. In light of the aforementioned challenges, we propose QuantumMedDx, a hybrid quantum–classical learning framework for classifying gliomas using MRI. The framework combines quantum feature encoding and variational quantum circuits with classical neural inference to improve diagnostic performance. The base model, QImageNet, uses amplitude-based quantum encoding for writing, entanglement-enabled parameterised quantum circuits (EPQCs) as feature extractors, and classical dense layers for classifying HGG and LGG from multimodal MRI slices. We demonstrate the effectiveness of the proposed approach on the BraTS 2021 benchmark dataset using a patient-aware 5-fold cross-validation protocol. Experimental results show that QuantumMedDx achieves accuracies of 94.12%, 93.30%, and 96.42%; F1-scores of 93.30% and 96.42%; and AUCs of 96.42% and 96.42%, respectively, outperforming conventional CNN, DNN, and SVM baselines. Ablation studies provide additional evidence of the performance improvements enabled by quantum Fourier transform and entanglement layers. Such results suggest that quantum–classical learning can efficiently improve feature extraction and discrimination in medical imaging, thus providing a modular and scalable route towards quantum–inspired clinical decision–support systems of the future.
Multiclass brain tumor classification from magnetic resonance imaging (MRI) remains challenging due to subtle inter-class visual similarities, particularly for anatomically small and heterogeneous tumor types such as pituitary lesions. To address these challenges, this paper presents a hybrid classical-quantum learning framework that integrates a deep convolutional feature extractor from the ResNet family with a variational quantum neural network (QNN) head for enhanced feature representation. In the present work, classical deep features extracted by the ResNet model are encoded into a six-qubit quantum circuit, enabling richer feature interactions within a high-dimensional Hilbert space. In addition, an anatomically guided Kernel-based Center Cropping (KCC) strategy is introduced as a deterministic spatial prior to emphasize anatomically relevant central brain structures, particularly benefiting challenging pituitary tumor cases. The proposed Kernel-based Center Cropping–Hybrid Quantum Neural Network (KCC-HQNN) framework is trained and evaluated on a publicly available composite four-class brain MRI dataset curated from Kaggle sources, comprising glioma, meningioma, pituitary tumor, and no-tumor categories. Experimental results across multiple ResNet variants demonstrate that the hybrid quantum–classical models consistently outperform their purely classical counterparts, achieving classification accuracy of up to 99%, while macro-averaged precision, recall, and F1 scores consistently fall in the range of approximately 0.95–0.99 across different ResNet variants. Reliability and reproducibility are further validated through stratified five-fold cross-validation and repeated experiments using multiple random initialization seeds, yielding a mean classification accuracy of 99.00 ± 0.08%, macro F1-score of 0.986 ± 0.11, and weighted F1-score of 0.989 ± 0.12, while cross-dataset evaluation on independent IEEE DataPort MRI datasets demonstrates robust generalization under dataset shift. All experiments are conducted on publicly available retrospective datasets using a quantum simulator. These results indicate that combining quantum feature processing with anatomically informed spatial processing improves multiclass discrimination of brain tumors.