A recurring claim in applied quantum machine learning [1],[2],[3]. The second is data scarcity relative to model capacity, which makes over-parameterized classical networks prone to memorizing scanner-specific artifacts rather than tumor morphology [4],[5],[6]. Convolutional and transfer-learning pipelines (ResNet, EfficientNet, VGG) remain the strongest classical baselines when training data is abundant [7], [8]. Quantum machine learning has been proposed as a partial answer to the capacity-versus-data problem [9]. In a variational quantum classifier, classical features are mapped onto qubit rotations or phases, and a parameterized circuit learns a decision boundary in the resulting Hilbert space [10], [11]. Because an n-qubit circuit spans a 2ⁿ-dimensional state space using only O(n) physical qubits and, for fixed depth, O(n) trainable parameters, a compact quantum classifier is argued to generalize more efficiently from small samples than a classical network with far more weights [12],[13]. Data re-uploading strengthens this claim: repeating the encoding block between variational layers turns even a single- qubit circuit into a universal approximator over the frequency spectrum the encoding induces [14],[15]. Whether this holds in practice is separate from whether it is theoretically available [18]. Published quantum medical-imaging classifiers typically report a single circuit, once, at a single qubit count, against classical baselines that are rarely varied over more than one architectural axis [16], [17]. It is therefore difficult to attribute a reported accuracy figure to quantum processing rather than to encoding, classifier family, qubit budget, or a favorable training run [19],[20]. This paper instead treats the quantum stage as an experimental platform, varying qubit count (8 vs. 16), encoding family (angle, phase, IQP/ZZ), entanglement topology (CZ-ring, CNOT-ring), and classifier family (QNN, QSVM, quantum ensemble, QCNN) one at a time against a fixed preprocessing and evaluation protocol [21],[22]. This design attributes an observed accuracy change to the pipeline component that produced it, and it lets us report honestly that a deeper, angle- encoded variational circuit does not converge reliably at the qubit
A recurring claim in applied quantum machine learning [1],[2],[3]. The second is data scarcity relative to model capacity, which makes over-parameterized classical networks prone to memorizing scanner-specific artifacts rather than tumor morphology [4],[5],[6]. Convolutional and transfer-learning pipelines (ResNet, EfficientNet, VGG) remain the strongest classical baselines when training data is abundant [7], [8]. Quantum machine learning has been proposed as a partial answer to the capacity-versus-data problem [9]. In a variational quantum classifier, classical features are mapped onto qubit rotations or phases, and a parameterized circuit learns a decision boundary in the resulting Hilbert space [10], [11]. Because an n-qubit circuit spans a 2ⁿ-dimensional state space using only O(n) physical qubits and, for fixed depth, O(n) trainable parameters, a compact quantum classifier is argued to generalize more efficiently from small samples than a classical network with far more weights [12],[13]. Data re-uploading strengthens this claim: repeating the encoding block between variational layers turns even a single- qubit circuit into a universal approximator over the frequency spectrum the encoding induces [14],[15]. Whether this holds in practice is separate from whether it is theoretically available [18]. Published quantum medical-imaging classifiers typically report a single circuit, once, at a single qubit count, against classical baselines that are rarely varied over more than one architectural axis [16], [17]. It is therefore difficult to attribute a reported accuracy figure to quantum processing rather than to encoding, classifier family, qubit budget, or a favorable training run [19],[20]. This paper instead treats the quantum stage as an experimental platform, varying qubit count (8 vs. 16), encoding family (angle, phase, IQP/ZZ), entanglement topology (CZ-ring, CNOT-ring), and classifier family (QNN, QSVM, quantum ensemble, QCNN) one at a time against a fixed preprocessing and evaluation protocol [21],[22]. This design attributes an observed accuracy change to the pipeline component that produced it, and it lets us report honestly that a deeper, angle- encoded variational circuit does not converge reliably at the qubit
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 and computationally efficient classification of brain tumors from MRI scans is critical for deployment in resource-constrained clinical settings, where conventional deep CNN architectures-despite achieving strong accuracy-often require large parameter counts that limit scalability. This work proposes a hybrid quantum-classical deep learning framework that combines a classical convolutional feature extractor with a variational quantum circuit (VQC) for multi-class tumor classification. Spatial features extracted by the CNN backbone will be dimensionality-reduced and encoded into quantum states using angle encoding; a parameterized quantum circuit will then process these features in a higher-dimensional Hilbert space, leveraging superposition and entanglement to capture non-linear feature correlations. Measurement outputs from the quantum layer will be passed to a classical fully connected layer for classification into glioma, meningioma, pituitary tumor, and no-tumor categories. The proposed framework will be evaluated on a publicly available benchmark MRI dataset and benchmarked against state-of-the-art classical CNN architectures in terms of accuracy, precision, recall, F1-score, and parameter efficiency. This study aims to demonstrate that quantum-enhanced layers can achieve competitive classification performance while substantially reducing trainable parameters, offering a practical direction for efficient, quantum-assisted medical image analysis.
Furthermore, the physics and multimodality imaging characteristics of tumors, along with high variability in intensity, morphology and anatomical position of the tumor, make the diagnosis of tumors from magnetic resonance imaging (MRI) still quite difficult. In this study, we introduce a hybrid quantum-classical Quantum Convolutional Neural Network (QCNN) that applies Quantum processing in the localized stage of feature extraction rather than later in the classification stage when most current quantum neural networks (CNNs) are used, for automated tumor detection from multiparametric MRI data. The proposed framework makes use of the BraTS 2021 data set, comprising of T1-weighted, post-contrast T1-weighted (T1Gd), T2-weighted and T2-FLAIR images, along with patient-level partitioning, intensity normalization, tumor relevant slice selection, 4×4/8×8 patch generation, dimensionality reduction, quantum angle encoding, trainable variational quantum transformations and Pauli-Z expectation-value measurements. This produced quantum feature maps which were summed across classical layers, and compared to a standard CNN and transfer learning CNN and CNN with a quantum classifier. The proposed QCNN showed excellent performance with 96% accuracy, 93% sensitivity, 94% specificity, 94% precision, 94% F1-score, AUPRC of 0.94, AUROC of 0.96 and the MCC score is 0.82, which is the best among all the architectures considered. The use of architectural analysis revealed to be an option, since 4-6 qubits, 2-3 quantum layers, angle encoding and linear or ring entanglement prove to be adequate. The results of robustness analysis showed that the performance is degrades progressively as the level of quantum noise is increased but still reached a good AUROC (0.80) in the case of strong quantum noise, and was confirmed by an independent external test with a value of 0.85. The results here show trainable quantum convolution to be one potential approach to localized multimodal representation learning for MRI, and present an indication of the impact of quantum noise, finite shot execution, and dataset shift. The study provides a groundwork for the further expansion of research into the use of quantum-enhanced MRI for analysis, in the form of an experimentally controlled experiments design, in larger independent cohorts of patients and with real quantum infrastructure.
This research work discusses three methods for detecting brain tumors using magnetic resonance imaging (MRI) - a pre-trained convolutional neural network (CNN) baseline, a pure quantum machine learning (QML) classifier that utilizes amplitude embedding and variational quantum circuits, and a hybrid of QML with CNN backbone model with a parameterised quantum circuit (PQC). All models were trained and tested on the four-class brain tumor MRI dataset (glioma, meningioma, pituitary, and no tumor) using the same preprocessing and evaluation protocols. The typical classical convolutional neural network (CNN) baseline achieved a test accuracy of 86.1%, with an F1 score of 0.8512. By comparison, the pure QML model provided a rather poor model, achieving limited accuracy of 44.7% and F1 of 0.4320, largely due to issues with shallow variational circuits and amplitude encoding on higher-dimensional data. The hybrid CNN-QML model provided the best results, achieving a test accuracy of 94.9% and an F1 of 0.9469, surpassing the typical classical baseline. These results indicate while standalone QML models are currently limited by the expressivity of quantum circuits and the scalability of their dimensionality, hybrid quantum-classical models can increase medical image classification performance, even when starting from a strong classical baseline.
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.
The characterization of brain tumors from magnetic resonance imaging (MRI) is crucial for diagnosis, treatment planning, and prognosis. But the analysis of brain tumors based on MRI is still difficult because of the significant intra-tumoral heterogeneity, ambiguous boundaries of lesions, high dimensionality of imaging features, and complicated nonlinear relationships between anatomical structures. While transformer-based deep learning (DL) models have greatly advanced automated tumor analysis, classical feature representations often fail to effectively capture complex feature interactions in pathological images, especially when distinguishing between subtle pathological variations and normal tissue needs to be done efficiently. Recent quantum machine learning (QML) developments indicate that quantum feature encoding can map complex data to exponentially larger Hilbert spaces, leading to a more expressive representation of nonlinear patterns and feature correlations that are challenging to model with conventional architectures. Encouraged by this potential, this study proposes a hybrid quantum-classical framework, named QFormer-Brain, which combines transformer-based segmentation with quantum representation learning for automated brain tumor segmentation and classification. To segment the tumor regions, a Shifted Window UNet Transformers (Swin-UNETR) architecture is first used to capture both local anatomical details and global contextual information through hierarchical self-attention. Then, the deep semantic features of the segmented lesions are mapped to the quantum feature mapping module with angle encoding and variational quantum circuits (VQCs), which can obtain richer nonlinear feature representation by quantum superposition and entanglement mechanisms. The obtained quantum embeddings are then combined with features obtained from the transformer and classified by a Quantum Transformer Classifier. The experiments were performed on the publicly available BraTS 2021 data set using an 8-qubit variational quantum circuit, with 1024 measurement shots and readout-error mitigation, via the IBM Qiskit Aer simulator. The proposed framework achieved a Dice Similarity Coefficient (DSC) of 97.1%, Intersection-over-Union (IoU) of 95.0%, classification accuracy of 98.5%, sensitivity of 98.1%, specificity of 99.0%, and F1-score of 98.3%, outperforming U-Net, nnU-Net, ResNet50, Vision Transformer (ViT), and conventional hybrid Convolutional Neural Network (CNN)-based models. These results show that quantum-inspired representation learning can be efficiently integrated into transformer-based models to boost the discriminative feature extraction from complex MRI data.