pith. machine review for the scientific record. sign in

arxiv: 2604.22877 · v1 · submitted 2026-04-24 · 🪐 quant-ph · cs.LG

Recognition: unknown

A Specialized Importance-Aware Quantum Convolutional Neural Network with Ring-Topology (IA-QCNN) for MGMT Promoter Methylation Prediction in Glioblastoma

Emine Akpinar, Murat Oduncuoglu

Authors on Pith no claims yet

Pith reviewed 2026-05-08 12:17 UTC · model grok-4.3

classification 🪐 quant-ph cs.LG
keywords quantum convolutional neural networkMGMT promoter methylationglioblastomaradiogenomicsmpMRIT1Gdquantum machine learningbiomarker prediction
0
0 comments X

The pith

A ring-topology quantum CNN predicts MGMT promoter methylation in glioblastoma patients with high accuracy using few parameters and less overfitting than classical models.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper sets out to demonstrate that quantum principles of superposition and entanglement can be harnessed in a convolutional network to extract useful features from complex, heterogeneous MRI scans for a key molecular marker in aggressive brain cancer. Classical deep learning approaches often overfit on such high-dimensional and spatially variable data, limiting their reliability for non-invasive biomarker prediction. The proposed architecture combines energy-based slice selection, importance-aware weighting, ring-structured quantum convolution, and folding pooling to achieve efficient representation learning in a high-dimensional space. If successful, this would offer a lighter, more robust alternative for guiding chemotherapy decisions based on MGMT status without tissue sampling. The work also finds that T1Gd images carry more predictive signal than multi-parametric sequences and that controlled noise can improve rather than harm results.

Core claim

The IA-QCNN model integrates energy-based slice selection, importance-aware weighting, ring-topology quantum convolution, and folding-based pooling to predict MGMT promoter methylation status from mpMRI and T1Gd images, achieving high accuracy with a low number of trainable parameters while reducing the overfitting typical of classical models and showing robustness when noise is present in the input data.

What carries the argument

The importance-aware quantum convolutional neural network with ring topology (IA-QCNN), which applies quantum convolution operations arranged in a ring structure together with importance weighting to perform feature extraction directly in high-dimensional Hilbert space.

If this is right

  • T1Gd images provide stronger discriminative information for MGMT status than standard multi-parametric MRI sequences.
  • Noise in the imaging data can function as a beneficial regularizer that improves predictive performance rather than degrading it.
  • Quantum convolutional layers with ring topology enable accurate biomarker prediction while using far fewer trainable parameters than classical equivalents.
  • The framework establishes a direct methodological connection between radiogenomic MRI analysis and quantum deep learning techniques.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Architectures of this form could be adapted to predict other molecular markers from imaging data in additional cancer types.
  • Lower parameter counts may allow such models to run on standard clinical workstations without specialized hardware.
  • Validation on larger, multi-institutional datasets would be required to assess whether the reported robustness holds across different scanners and patient populations.
  • Hybrid quantum-classical pipelines might further scale the approach for real-time clinical decision support.

Load-bearing premise

The specific combination of energy-based slice selection, importance-aware weighting, ring-topology quantum convolution, and folding pooling produces more efficient feature learning than classical neural networks when applied to correlated, high-dimensional radiogenomic MRI data.

What would settle it

A classical convolutional neural network trained and tested on the identical mpMRI and T1Gd dataset achieving equal or higher accuracy with a comparable or lower parameter count would falsify the claimed efficiency advantage of the quantum architecture.

Figures

Figures reproduced from arXiv: 2604.22877 by Emine Akpinar, Murat Oduncuoglu.

Figure 1
Figure 1. Figure 1: General workflow of the proposed study. In the first stage, the RSNA-MICCAI Brain Tumor Radiogenomic dataset and the mpMRI sequences employed in this study are introduced. The second stage outlines the DICOM preprocessing, normalization, and rescaling steps applied to the mpMRI images, with particular emphasis on T1Gd MRI, alongside the proposed “Energy-Based Slice Selection” method. In the third stage, a … view at source ↗
Figure 2
Figure 2. Figure 2: Distribution of the RSNA–MICCAI Brain Tumor Radiogenomic dataset according to various characteristics. (a) Class distribution based on MGMT status, comprising 307 cases with positive (methylated) MGMT promoter status and 278 cases with negative (unmethylated) status. (b) Total number of slices calculated across all DICOM files for each MRI modality, with 74,248, 77,627, 96,766, and 100,000 slices for FLAIR… view at source ↗
Figure 3
Figure 3. Figure 3: Examples of mpMRI modalities from two randomly selected cases with positive and negative MGMT promoter methylation status. Panels (a–d) depict FLAIR, T1w, T1Gd, and T2 images from an MGMT-methylated (positive) case, respectively, while panels (e–h) illustrate FLAIR, T1w, T1Gd, and T2 images from an MGMT￾unmethylated (negative) case view at source ↗
Figure 4
Figure 4. Figure 4: Sequential representation of T1Gd MRI slices obtained from Case 146 in the dataset. From the series of 76 slices, images with low information content were filtered out, and only slices containing meaningful anatomical structures were selected for visualization. 2.2.Image Preprocessing Following the dataset definition, all DICOM images corresponding to the mpMRI modalities were subjected to normalization to… view at source ↗
Figure 5
Figure 5. Figure 5: illustrates the workflow of the proposed energy-based slice selection method for a representative GBM patient, while view at source ↗
Figure 6
Figure 6. Figure 6: The top K = 10 slices selected based on slice-wise energy scores computed using the proposed energy￾based slice selection method, along with their corresponding T1Gd MRI representations. 2.3.Specialized Importance-Aware Quantum Convolutional Neural Network with Ring￾Topology (IA-QCNN) Architecture view at source ↗
Figure 7
Figure 7. Figure 7: Pseudocode of the proposed algorithm, including the steps for normalizing input features via angle transformation, implementing angle-based quantum feature encoding, and applying the learnable importance￾aware weighting mechanism. 2.3.1.1. Angle Transformation Following the selection of ten T1Gd slices per patient via the Energy-Based Slice Selection method, each MRI image is subsequently processed using a… view at source ↗
Figure 8
Figure 8. Figure 8: Quantum circuit diagram of the proposed angle-based quantum feature encoding and importance-aware weighting strategy, implemented using the TensorFlow Quantum (TFQ) and Cirq frameworks. For visual clarity, the circuit diagram is representatively illustrated using 8 qubits. In the actual experiments, the number of qubits was automatically determined based on the dimension obtained after PCA and constrained … view at source ↗
Figure 9
Figure 9. Figure 9: Quantum circuit diagram of the proposed ring-topology quantum convolution and folding-based quantum pooling approaches, implemented using the TensorFlow Quantum (TFQ) and Cirq frameworks. Due to space constraints, the circuit diagram is representatively illustrated for 8 qubits. The complete diagrams for the quantum circuits generated for all automatically determined qubit counts are provided in Supplement… view at source ↗
Figure 10
Figure 10. Figure 10: Epoch-wise variation of slice-level accuracy and loss values for the prediction of MGMT promoter methylation status in GBM tumors using T1Gd MRI images via the IA-QCNN architecture. In the subsequent stage, the slice-based prediction probabilities obtained for each patient were combined using a “patient-level approach”, in alignment with the clinical decision-making process. At this stage, the probability… view at source ↗
Figure 12
Figure 12. Figure 12: Receiver Operating Characteristic (ROC) curve obtained by aggregating slice-level IA-QCNN predictions derived from T1Gd MRI images into a patient-level representation for predicting MGMT promoter methylation status in GBM tumors. An AUC score of 0.66 was achieved. As a result of aggregating slice-level predictions into patient-level representations via mean averaging for each patient, the precision, recal… view at source ↗
Figure 13
Figure 13. Figure 13: Detailed representation of precision, recall, and F1-score values for methylated (1) and unmethylated (0) classes according to MGMT promoter methylation status. 3.2.Results of the specialized IA-QCNN architecture in predicting MGMT promoter methylation from mpMRI images At this stage of the study, the prediction of MGMT promoter methylation status in GBM tumors was performed using the proposed IA-QCNN arc… view at source ↗
Figure 14
Figure 14. Figure 14: Epoch-wise variation of slice-level accuracy and loss values for the prediction of MGMT promoter methylation status in GBM tumors using mpMRI images via the IA-QCNN architecture. As implemented for the T1Gd MRI modality in the previous stage, a mean aggregation method was employed here to transition from slice-level results to a patient-level approach. Following the testing phase, the overall patient-leve… view at source ↗
Figure 15
Figure 15. Figure 15: Patient-level confusion matrix for the prediction of MGMT promoter methylation status in GBM tumors derived from mpMRI images view at source ↗
Figure 16
Figure 16. Figure 16: ROC curve and corresponding AUC value (AUC = 0.45) obtained by aggregating slice-level IA-QCNN predictions into a patient-level representation for MGMT promoter methylation in GBM tumors using mpMRI images. As a result of the patient-level mean aggregation of slice-based predictions performed with mpMRI modalities for each patient, the precision, recall, and F1-score values for the methylated class were o… view at source ↗
Figure 17
Figure 17. Figure 17: Detailed representation of class-wise precision, recall, and F1-score values obtained by aggregating slice-level predictions of the specialized IA-QCNN model into patient-level representation using mpMRI modalities. 3.3.Results of the specialized IA-QCNN architecture under mild-to-moderate Gaussian noise in T1Gd MRI images In this phase of the study, Gaussian noise was added to all mpMRI images of GBM tum… view at source ↗
Figure 18
Figure 18. Figure 18: Epoch-wise variation of slice-level training and validation accuracy/loss values for the IA-QCNN architecture in predicting MGMT promoter methylation status under the addition of Gaussian noise to T1Gd MRI images. Additionally, for the Gaussian-noise-perturbed T1Gd MRI images, a mean aggregation method was employed to transition from slice-level results to a patient-level approach. As a result of this hie… view at source ↗
Figure 19
Figure 19. Figure 19: Patient-level confusion matrix of the IA-QCNN architecture evaluated on Gaussian-noise-perturbed T1Gd MRI images for predicting MGMT promoter methylation status view at source ↗
Figure 20
Figure 20. Figure 20: ROC curve and corresponding AUC value (AUC = 0.67) obtained by aggregating slice-level predictions into patient-level assessments for the IA-QCNN architecture evaluated on Gaussian-noise-perturbed T1Gd MRI images. As a result of aggregating slice-level predictions obtained from Gaussian-noise-perturbed T1Gd MRI modalities into patient-level representations via mean averaging for each patient, the precisio… view at source ↗
Figure 21
Figure 21. Figure 21: Detailed representation of class-wise precision, recall, and F1-score values, obtained by the patient￾level mean aggregation of slice-level predictions from the specialized IA-QCNN model using Gaussian-noise￾perturbed T1Gd MRI modalities. 3.4.Results of the specialized IA-QCNN architecture under gate-level noise in the quantum circuit In this section, Gaussian noise was introduced to the parameters of the… view at source ↗
Figure 22
Figure 22. Figure 22: Epoch-wise variation of slice-level training and validation accuracy and loss values for the IA-QCNN architecture in predicting MGMT promoter methylation status under the addition of Gaussian noise to the rotational gate parameters within the quantum circuit. Furthermore, when transitioning from slice-level results to patient-level outcomes, the overall patient￾level accuracy of the IA-QCNN model with Gau… view at source ↗
Figure 23
Figure 23. Figure 23: , and the ROC curve is provided in view at source ↗
Figure 25
Figure 25. Figure 25: Detailed representation of class-wise precision, recall, and F1-score values, obtained by the patient￾level mean aggregation of slice-level predictions from the customized IA-QCNN model under the addition of Gaussian noise to the rotational gate parameters within the quantum circuit. 3.5.Results of the specialized IA-QCNN architecture under hybrid noise (multi-level Gaussian perturbations) in both T1Gd MR… view at source ↗
Figure 26
Figure 26. Figure 26: Epoch-wise variation of slice-level training and validation accuracy and loss for the specialized IA￾QCNN architecture in predicting MGMT promoter methylation status under the designed hybrid noise scenario (multi-level Gaussian perturbations), where noise was applied simultaneously to both T1Gd MRI images and rotational gate blocks within the quantum circuit. Furthermore, when transitioning from slice-le… view at source ↗
Figure 27
Figure 27. Figure 27: Patient-level confusion matrix of the specialized IA-QCNN architecture under the hybrid noise scenario (multi-level Gaussian perturbations) for MGMT promoter methylation status prediction view at source ↗
Figure 28
Figure 28. Figure 28: ROC curve and AUC value (AUC = 0.75) of the specialized IA-QCNN architecture under the hybrid noise scenario (multi-level Gaussian perturbations), obtained by aggregating slice-level predictions on a patient￾level basis. Under the hybrid noise scenario, the slice-level predictions of the specialized IA-QCNN architecture were aggregated into patient-level outcomes via averaging, yielding precision, recall,… view at source ↗
Figure 29
Figure 29. Figure 29: Detailed representation of class-wise precision, recall, and F1-score values for the specialized IA￾QCNN architecture under the hybrid noise scenario (multi-level Gaussian perturbations). 3.6.Comparative results of CNN and DNN models designed with a similar number of parameters, along with Transfer Learning (TL) and State-of-the-Art (SOTA) models In this stage of the study, the performance of the proposed… view at source ↗
Figure 30
Figure 30. Figure 30: Epoch-wise variation of training and validation accuracy and loss for the DNN architecture designed with a similar number of parameters to the specialized IA-QCNN model for performance comparison. In the subsequent stage, as in quantum-based approaches, slice-level prediction probabilities were aggregated into a patient-level representation in accordance with clinical decision-making requirements. As a re… view at source ↗
Figure 31
Figure 31. Figure 31: Patient-level confusion matrix of the DNN architecture for the prediction of MGMT promoter methylation status view at source ↗
Figure 32
Figure 32. Figure 32: ROC curve and AUC value (AUC = 0.69) for the DNN architecture, obtained by aggregating slice￾level predictions on a patient-level basis. Furthermore, upon the patient-level mean aggregation of slice-level results obtained during the MGMT promoter methylation status prediction process of the DNN architecture, the precision, recall, and F1- score for the methylated class were recorded as 0.64, 0.72, and 0.6… view at source ↗
Figure 33
Figure 33. Figure 33: Epoch-wise variation of training and validation accuracy and loss for the CNN architecture designed with a similar number of parameters to the specialized IA-QCNN model for performance comparison. Furthermore, upon transitioning from slice-level to patient-level results, the overall patient-level accuracy of the CNN architecture on the test set was recorded as 0.61. The patient-level confusion view at source ↗
Figure 34
Figure 34. Figure 34: Patient-level confusion matrix of the CNN architecture for MGMT promoter methylation prediction view at source ↗
Figure 35
Figure 35. Figure 35: ROC curve and corresponding AUC value (AUC = 0.65) obtained by aggregating slice-level predictions of the CNN architecture into a patient-level representation. In the patient-level evaluation of the CNN architecture, the precision, recall, and F1-score for the methylated class were recorded as 0.59, 0.79, and 0.68, respectively; for the unmethylated class, these values were determined to be 0.64, 0.41, an… view at source ↗
read the original abstract

GBM is a highly aggressive primary malignancy in adults, necessitating personalized therapeutic strategies due to its inherent molecular heterogeneity. MGMT promoter methylation is a pivotal prognostic biomarker for anticipating response to temozolomide-based chemotherapy. Although various AI frameworks have been developed for non-invasive MGMT prediction, spatial heterogeneity of methylation status and the high-dimensional and correlated nature of MRI data frequently constrain discriminative feature learning and generalizability of classical models. To circumvent these limitations, a specialized IA-QCNN architecture is proposed, based on the principles of quantum mechanics, including superposition and entanglement, and enabling more efficient representation learning in high-dimensional Hilbert space. The framework establishes a methodological bridge between GBM radiogenomics and quantum deep learning by integrating energy-based slice selection, importance-aware weighting, ring-topology quantum convolution, and folding-based pooling layers. When the model predicts MGMT promoter methylation status using both mpMRI and T1Gd images, experimental results demonstrate that the IA-QCNN achieves high accuracy despite its low number of trainable parameters while effectively minimizing the overfitting problem observed in classical models. Quantitative analyses reveal that the T1Gd modality possesses higher discriminative power than mpMRI, establishing a clinically significant sequence preference. Furthermore, the model exhibits exceptional robustness in hybrid noise environments, effectively utilizing noise as a regularization mechanism to enhance predictive performance. Consequently, the specialized IA-QCNN architecture provides a robust and computationally efficient alternative to classical approaches in the analysis of heterogeneous radiogenomic data.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript introduces an Importance-Aware Quantum Convolutional Neural Network with Ring-Topology (IA-QCNN) for non-invasive prediction of MGMT promoter methylation status in glioblastoma multiforme (GBM) from multi-parametric MRI (mpMRI) and T1Gd images. The architecture combines energy-based slice selection, importance-aware weighting, ring-topology quantum convolution, and folding-based pooling layers to exploit superposition and entanglement for representation learning in high-dimensional Hilbert space. The central claims are that IA-QCNN achieves high accuracy with a low number of trainable parameters, reduces overfitting relative to classical CNNs, demonstrates superior discriminative power for the T1Gd modality, and exhibits robustness to hybrid noise that can act as regularization.

Significance. If the performance and generalization claims are substantiated with rigorous validation, the work would represent a concrete demonstration of quantum machine learning advantages in radiogenomics, particularly for handling spatial heterogeneity and high-dimensional correlated MRI data with reduced parameter count and overfitting. The noise-robustness finding and modality preference for T1Gd could have direct clinical implications for efficient biomarker prediction.

major comments (3)
  1. [Experimental Results / Validation Procedure] The experimental validation does not specify patient-level cross-validation or partitioning. Because energy-based slice selection and folding-based pooling operate at the slice level on heterogeneous GBM volumes, intra-patient slice leakage across folds is possible; this risks inflated accuracy metrics that reflect patient-specific imaging artifacts rather than methylation-specific radiogenomic features, directly undermining the claim that the ring-topology QCNN plus importance weighting yields genuinely superior Hilbert-space representations.
  2. [Abstract and Results] The abstract asserts 'high accuracy' and 'minimizing the overfitting problem' without reporting numerical values, baseline comparisons (e.g., classical CNN accuracies), dataset sizes (patients/slices), error bars, or metrics such as AUC or sensitivity. If these details are absent or insufficiently detailed in the full results section, the central performance claims remain unsupported by evidence.
  3. [Methods / Importance-Aware Weighting] The importance-aware weighting introduces additional free parameters (listed among the model's trainable components). It is unclear whether these weights are optimized on the same data used for final accuracy reporting or held out, raising the possibility of circular performance claims that would invalidate the assertion of reduced overfitting and parameter efficiency.
minor comments (2)
  1. [Abstract] The abstract and introduction would benefit from explicit statements of the number of patients, total slices, and train/validation/test split ratios to allow immediate assessment of statistical power.
  2. [Methods] Notation for the quantum circuit parameters and the ring-topology convolution operator should be defined consistently with standard quantum information conventions (e.g., explicit reference to the number of qubits and entanglement structure) to improve reproducibility.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their thorough and constructive review. We address each major comment point by point below, with revisions made where the manuscript required clarification or expansion to strengthen the presentation of our methods and results.

read point-by-point responses

  1. Referee: [Experimental Results / Validation Procedure] The experimental validation does not specify patient-level cross-validation or partitioning. Because energy-based slice selection and folding-based pooling operate at the slice level on heterogeneous GBM volumes, intra-patient slice leakage across folds is possible; this risks inflated accuracy metrics that reflect patient-specific imaging artifacts rather than methylation-specific radiogenomic features, directly undermining the claim that the ring-topology QCNN plus importance weighting yields genuinely superior Hilbert-space representations.

    Authors: We agree that explicit patient-level partitioning is necessary to rule out slice leakage in multi-slice GBM imaging data. Our experiments did employ patient-wise assignment of slices to folds during cross-validation. However, this was not stated clearly enough in the original manuscript. We have revised the Experimental Setup section to describe the patient-level partitioning procedure in detail, including confirmation that no slices from the same patient cross folds. This addition directly mitigates the concern and supports the validity of the reported performance. revision: yes

  2. Referee: [Abstract and Results] The abstract asserts 'high accuracy' and 'minimizing the overfitting problem' without reporting numerical values, baseline comparisons (e.g., classical CNN accuracies), dataset sizes (patients/slices), error bars, or metrics such as AUC or sensitivity. If these details are absent or insufficiently detailed in the full results section, the central performance claims remain unsupported by evidence.

    Authors: The referee is correct that the abstract used qualitative phrasing without accompanying numbers. While quantitative results, baselines, dataset sizes, and metrics appear in the Results section, we have updated the abstract to report specific accuracy values, classical CNN comparisons, patient and slice counts, error bars, AUC, and sensitivity. We have also verified that the Results section already contains these details with appropriate statistical reporting. This revision makes the central claims fully supported by evidence in both abstract and body. revision: yes

  3. Referee: [Methods / Importance-Aware Weighting] The importance-aware weighting introduces additional free parameters (listed among the model's trainable components). It is unclear whether these weights are optimized on the same data used for final accuracy reporting or held out, raising the possibility of circular performance claims that would invalidate the assertion of reduced overfitting and parameter efficiency.

    Authors: We appreciate the referee pointing out this ambiguity in the description of the trainable importance weights. These parameters are optimized jointly with the rest of the model but strictly within each cross-validation training fold using only the training split; they are never tuned on held-out test data. We have expanded the Methods section to explicitly document this training protocol and the separation of optimization from evaluation. The revision removes any possibility of circularity and reinforces the claims of parameter efficiency and overfitting reduction. revision: yes

Circularity Check

0 steps flagged

No circularity; architecture proposal and experimental claims are independent of self-referential fitting

full rationale

The paper proposes an IA-QCNN model with components like energy-based slice selection, importance-aware weighting, ring-topology quantum convolution, and folding-based pooling. The central claim rests on reported experimental accuracy for MGMT prediction using mpMRI and T1Gd data, with assertions of low parameter count and reduced overfitting versus classical models. No equations, derivations, or self-citations are present in the text that define a result in terms of itself, rename a fitted parameter as a prediction, or import uniqueness via author-overlapping citations. The importance-aware elements are described as design choices enabling Hilbert-space representation, not as post-hoc fits to the same evaluation metrics. Experimental results are presented as validation rather than tautological outputs. The derivation chain is self-contained as a methodological proposal backed by empirical testing.

Axiom & Free-Parameter Ledger

2 free parameters · 1 axioms · 0 invented entities

The central claim rests on unverified assumptions that quantum-inspired layers will outperform classical ones on MRI data and that the listed architectural components can be realized without prohibitive simulation cost. No independent evidence for these assumptions is supplied in the abstract.

free parameters (2)
  • importance weights
    The importance-aware weighting mechanism requires parameters that are almost certainly fitted to training data.
  • quantum circuit parameters
    Trainable parameters inside the ring-topology quantum convolution layers are fitted during optimization.
axioms (1)
  • domain assumption Quantum superposition and entanglement enable more efficient representation learning in high-dimensional Hilbert space for correlated MRI data.
    Explicitly invoked in the abstract as the foundational principle of the IA-QCNN framework.

pith-pipeline@v0.9.0 · 5570 in / 1406 out tokens · 62783 ms · 2026-05-08T12:17:51.002357+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

94 extracted references · 91 canonical work pages

  1. [1]

    Singh, D

    S. Singh, D. Dey, D. Barik, I. Mohapatra, S. Kim, M. Sharma, S. Prasad, P. Wang, A. Singh, G. Singh, Glioblastoma at the crossroads: current understanding and future therapeutic horizons, Sig Transduct Target Ther 10 (2025) 213. https://doi.org/10.1038/s41392-025-02299-4

    work page doi:10.1038/s41392-025-02299-4 2025

  2. [2]

    Pouyan, M

    A. Pouyan, M. Ghorbanlo, M. Eslami, M. Jahanshahi, E. Ziaei, A. Salami, K. Mokhtari, K. Shahpasand, N. Farahani, T.E. Meybodi, M. Entezari, A. Taheriazam, K. Hushmandi, M. Hashemi, Glioblastoma multiforme: insights into pathogenesis, key signaling pathways, and therapeutic strategies, Mol Cancer 24 (2025) 58. https://doi.org/10.1186/s12943-025-02267-0

    work page doi:10.1186/s12943-025-02267-0 2025

  3. [3]

    Holland, Glioblastoma multiforme: The terminator, Proc

    E.C. Holland, Glioblastoma multiforme: The terminator, Proc. Natl. Acad. Sci. U.S.A. 97 (2000) 6242 –6244. https://doi.org/10.1073/pnas.97.12.6242

    work page doi:10.1073/pnas.97.12.6242 2000

  4. [4]

    D. Roda, P. Veiga, J.B. Melo, I.M. Carreira, I.P. Ribeiro, Principles in the Management of Glioblastoma, Genes 15 (2024) 501. https://doi.org/10.3390/genes15040501

    work page doi:10.3390/genes15040501 2024

  5. [5]

    Möller, M

    C. Möller, M. Schoof , K.L. Ligon, U. Schüller, Integration of omics data in the diagnosis and therapy of glioblastoma, Brain Pathology 36 (2026) e70027. https://doi.org/10.1111/bpa.70027

    work page doi:10.1111/bpa.70027 2026

  6. [6]

    Ostrom, M

    Q.T. Ostrom, M. Price, C. Neff, G. Cioffi, K.A. Waite, C. Kruchko, J.S. Barnholtz -Sloan, CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2015–2019, Neuro-Oncology 24 (2022) v1–v95. https://doi.org/10.1093/neuonc/noac202

    work page doi:10.1093/neuonc/noac202 2015

  7. [7]

    Obrador, P

    E. Obrador, P. Moreno-Murciano, M. Oriol-Caballo, R. López-Blanch, B. Pineda, J. Gutiérrez-Arroyo, A. Loras, L. Gonzalez-Bonet, C. Martinez-Cadenas, J. Estrela, M. Marqués-Torrejón, Glioblastoma Therapy: Past, Present and Future, IJMS 25 (2024) 2529. https://doi.org/10.3390/ijms25052529

    work page doi:10.3390/ijms25052529 2024

  8. [8]

    Grech, T

    N. Grech, T. Dalli, S. Mizzi, L. Meilak, N. Calleja, A. Zrinzo, Rising Incidence of Glioblastoma Multiforme in a Well-Defined Population, Cureus (2020). https://doi.org/10.7759/cureus.8195

    work page doi:10.7759/cureus.8195 2020

  9. [9]

    Aldape, G

    K. Aldape, G. Zadeh, S. Mansouri, G. Reifenberger, A. Von Deimling, Glioblastoma: pat hology, molecular mechanisms and markers, Acta Neuropathol 129 (2015) 829–848. https://doi.org/10.1007/s00401-015-1432-1

    work page doi:10.1007/s00401-015-1432-1 2015

  10. [10]

    Tan, D.M

    A.C. Tan, D.M. Ashley, G.Y. López, M. Malinzak, H.S. Friedman, M. Khasraw, Management of glioblastoma: State of the art and future di rections, CA A Cancer J Clinicians 70 (2020) 299 –312. https://doi.org/10.3322/caac.21613

    work page doi:10.3322/caac.21613 2020

  11. [11]

    De Vleeschouwer, eds., Glioblastoma, Codon Publications, 2017

    Department of Neurosurgery, University Hospitals Leuven, Leuven, Belgium, S. De Vleeschouwer, eds., Glioblastoma, Codon Publications, 2017. https://doi.org/10.15586/codon.glioblastoma.2017

    work page doi:10.15586/codon.glioblastoma.2017 2017

  12. [12]

    Urbańska, J

    K. Urbańska, J. Sokołowska, M. Szmidt, P. Sysa, Review Glioblastoma multiforme – an overview, Wo 5 (2014) 307–312. https://doi.org/10.5114/wo.2014.40559

    work page doi:10.5114/wo.2014.40559 2014

  13. [13]

    Bleeker, R.J

    F.E. Bleeker, R.J. Molenaar, S. Leenstra, Recent advances in the molecular understanding of glioblastoma, J Neurooncol 108 (2012) 11–27. https://doi.org/10.1007/s11060-011-0793-0

    work page doi:10.1007/s11060-011-0793-0 2012

  14. [14]

    Davis, Glioblastoma: Overview of Disease and Treatment, CJON 20 (2016) S2 –S8

    M. Davis, Glioblastoma: Overview of Disease and Treatment, CJON 20 (2016) S2 –S8. https://doi.org/10.1188/16.CJON.S1.2-8

    work page doi:10.1188/16.cjon.s1.2-8 2016

  15. [15]

    Komori, The 2016 WHO Clas sification of Tumours of the Central Nervous System: The Major Points of Revision, Neurol

    T. Komori, The 2016 WHO Clas sification of Tumours of the Central Nervous System: The Major Points of Revision, Neurol. Med. Chir.(Tokyo) 57 (2017) 301–311. https://doi.org/10.2176/nmc.ra.2017-0010

    work page doi:10.2176/nmc.ra.2017-0010 2016

  16. [16]

    Louis, Arie Perry, Pieter Wesseling, Daniel J

    D.N. Louis, A. Perry, P. Wesseling, D.J. Brat, I.A. Cree, D. Figarella -Branger, C. Ha wkins, H.K. Ng, S.M. Pfister, G. Reifenberger, R. Soffietti, A. Von Deimling, D.W. Ellison, The 2021 WHO Classification of Tumors of the Central Nervous System: a summary, Neuro -Oncology 23 (2021) 1231 –1251. https://doi.org/10.1093/neuonc/noab106

    work page doi:10.1093/neuonc/noab106 2021

  17. [17]

    Diffuse astrocytic glioma, IDH-wildtype, with molecular features of glioblastoma, WHO grade IV,

    D.J. Brat, K. Aldape, H. Colman, E.C. Holland, D.N. Louis, R.B. Jenkins, B.K. Kleinschmidt -DeMasters, A. Perry, G. Reifenberger, R. Stupp, A. Von Deimling, M. Weller, cIMPACT -NOW update 3: recommended diagnostic criteria for “Diffuse astrocytic glioma, IDH-wildtype, with molecular features of glioblastoma, WHO grade IV,” Acta Neuropathol 136 (2018) 805–...

    work page doi:10.1007/s00401-018-1913-0 2018

  18. [18]

    Kinslow, A.I

    C.J. Kinslow, A.I. Rae, K. Taparra, P. Kumar, M.D. Siegelin, J. Grinband, B.J.A. Gill, G.M. McKhann, M.B. Sisti, J.N. Bruce, P.D. Canoll, F.M. Iwamoto, D.P. Horowitz, L.A. Kachnic, A.I. Neugut, J.B. Yu, S.K. Cheng, T.J.C. Wang, MGMT Promoter Methylation Predicts Overall Survival after Chemotherapy for 1p/19q-Codeleted Gliomas, Clinical Cancer Research 29 ...

    work page doi:10.1158/1078-0432.ccr-23-1295 2023

  19. [19]

    Crespo, A.L

    I. Crespo, A.L. Vital, M. Gonzalez-Tablas, M.D.C. Patino, A. Otero, M.C. Lopes, C. De Oliveira, P. Domingues, A. Orfao, M.D. Tabernero, Molecular and Genomic Alterations in Glioblastoma Multiforme, The American Journal of Pathology 185 (2015) 1820–1833. https://doi.org/10.1016/j.ajpath.2015.02.023

    work page doi:10.1016/j.ajpath.2015.02.023 2015

  20. [20]

    W. Wick, M. Weller, M. Van Den Bent, M. Sanson, M. Weiler, A. Von Deimling, C. Plass, M. Hegi, M. Platten, G. Reifenberger, MGMT testing —the challenges for b iomarker-based glioma treatment, Nat Rev Neurol 10 (2014) 372–385. https://doi.org/10.1038/nrneurol.2014.100

    work page doi:10.1038/nrneurol.2014.100 2014

  21. [21]

    Gerson, Clinical Relevance of MGMT in the Treatment of Cancer, JCO 20 (2002) 2388 –2399

    S.L. Gerson, Clinical Relevance of MGMT in the Treatment of Cancer, JCO 20 (2002) 2388 –2399. https://doi.org/10.1200/JCO.2002.06.110

    work page doi:10.1200/jco.2002.06.110 2002

  22. [22]

    Gerson, MGMT: its role in cancer aetiology and cancer therapeutics, Nat Rev Cancer 4 (2004) 296 –307

    S.L. Gerson, MGMT: its role in cancer aetiology and cancer therapeutics, Nat Rev Cancer 4 (2004) 296 –307. https://doi.org/10.1038/nrc1319

    work page doi:10.1038/nrc1319 2004

  23. [23]

    Stupp, M.E

    R. Stupp, M.E. Hegi, W.P. Mason, M.J. Van Den Bent, M.J. Taphoorn, R.C. Janzer, S.K. Ludwin, A. Allgeier, B. Fisher, K. Belanger, P. Hau, A.A. Brandes, J. Gijtenbeek, C. Marosi, C.J. Vecht, K. Mokhtari, P. Wesseling, S. Villa, E. Eisenhauer, T. Gorlia, M. Weller, D. Lacombe, J.G. Cairncross, R. -O. Mirimanoff, Effects of radiotherapy with concomitant and ...

    work page doi:10.1016/s1470-2045(09)70025-7 2009

  24. [24]

    Malmström, B.H

    A. Malmström, B.H. Grønberg, C. Marosi, R. Stupp, D. Frappaz, H. Schultz, U. Abacioglu, B. Tavelin, B. Lhermitte, M.E. Hegi, J. Rosell, R. Henriksson, Temozolomide versus standard 6 -week radiotherapy versus hypofractionated radiotherapy in patients older than 60 years with glioblastoma: the Nordic randomised, phase 3 trial, The Lancet Oncology 13 (2012) ...

    work page doi:10.1016/s1470-2045(12)70265-6 2012

  25. [25]

    W. Wick, M. Platten, C. Meisner, J. Felsberg, G. Tabatabai, M. Simon, G. Nikkhah, K. Papsdorf, J.P. Steinbach, M. Sabel, S.E. Combs, J. Vesper, C. Braun, J. Meixensberger, R. Ketter, R. Mayer-Steinacker, G. Reifenberger, M. Weller, Temozolomide chemotherapy alone versus radiotherapy alone for malignant astrocytoma in the elderly: the NOA -08 randomised, p...

    work page doi:10.1016/s1470-2045(12)70164-x 2012

  26. [26]

    J. Tang, N. Karbhari, J.L. Campian, Therapeutic Targets in Glioblastoma: Molecular Pathways, Emerging Strategies, and Future Directions, Cells 14 (2025) 494. https://doi.org/10.3390/cells14070494

    work page doi:10.3390/cells14070494 2025

  27. [27]

    Hegi, A.-C

    M.E. Hegi, A.-C. Diserens, S. Godard, P. -Y. Dietrich, L. Regli, S. Ostermann, P. Otten, G. Van Melle, N. De Tribolet, R. Stupp, Clinical Trial Substantiates the Predictive Value of O-6-Methylguanine-DNA Methyltransferase Promoter Methylation in Glioblastoma Patients Treated with Temozolomide, Clinical Cancer Research 10 (2004) 1871–1874. https://doi.org/...

    work page doi:10.1158/1078-0432.ccr-03-0384 2004

  28. [28]

    Korfiatis, T.L

    P. Korfiatis, T.L. Kline, L. Coufalova, D.H. Lachance, I.F. Parney, R.E. Carter, J.C. Buckner, B.J. Erickson, MRI texture features as biomarkers to predict MGMT methylation status in glioblastomas, Medical Physics 43 (2016) 2835–2844. https://doi.org/10.1118/1.4948668

    work page doi:10.1118/1.4948668 2016

  29. [29]

    Choi, S.K

    C. Choi, S.K. Ganji, R.J. DeBerardinis, K.J. Hatanpaa, D. Rakheja, Z. Kovacs, X. -L. Yang, T. Mashimo, J.M. Raisanen, I. Marin-Valencia, J.M. Pascual, C.J. Madden, B.E. Mickey, C.R. Malloy, R.M. Bachoo, E.A. Maher, 2-hydroxyglutarate detection by magnetic resonance spectroscopy in IDH -mutated patients with gliomas, Nat Med 18 (2012) 624–629. https://doi....

    work page doi:10.1038/nm.2682 2012

  30. [30]

    Sipos, B.L

    D. Sipos, B.L. Raposa, O. Freihat, M. Simon, N. Mekis, P. Cornacchione, Á. Kovács, Glioblastoma: Clinical Presentation, Multidisciplinary Management, and Long -Term Outcomes, Cancers 17 (2025) 146. https://doi.org/10.3390/cancers17010146

    work page doi:10.3390/cancers17010146 2025

  31. [31]

    D.-S. Kong, J. Kim, G. Ryu, H.-J. You, J.K. Sung, Y.H. Han, H.-M. Shin, I.-H. Lee, S.-T. Kim, C.-K. Park, S.H. Choi, J.W. Choi, H.J. Seol, J. -I. Lee, D. -H. Nam, Quantitative radiomic profiling of glioblastoma represents transcriptomic expression, Oncotarget 9 (2018) 6336–6345. https://doi.org/10.18632/oncotarget.23975

    work page doi:10.18632/oncotarget.23975 2018

  32. [32]

    Minh, Q.H

    T.N.T. Minh, Q.H. Kha, V.H. Le, M.C.H. Chua, MGMT Promoter Methylation Prediction in Glioblastoma Using 3D CNNs with Advanced MRI Sequences, (n.d.). https://openreview.net/pdf?id=CS7AhWnVnO

  33. [33]

    Chilaca -Rosas, M.T

    M.F. Chilaca -Rosas, M.T. Contreras -Aguilar, F. Pallach -Loose, N.F. Altamirano -Bustamante, D.R. Salazar - Calderon, C. Revilla -Monsalve, J.C. Heredia -Gutiérrez, B. Conde -Castro, R. Medrano -Guzmán, M.M. Altamirano-Bustamante, Systematic review and epistemic meta-analysis to advance binomial AI -radiomics integration for predicting high-grade glioma ...

    work page doi:10.1038/s41598-025-98058-0 2025

  34. [34]

    Halloum, H

    K. Halloum, H. Ez -Zahraouy, Advancing brain tumo ur segmentation: A novel CNN approach with Resnet50 and DrvU-Net: A comparative study, IDT 18 (2024) 2079–2096. https://doi.org/10.3233/IDT-240385

    work page doi:10.3233/idt-240385 2024

  35. [35]

    Naeem, O

    A.B. Naeem, O. Osman, S. Alsubai, T. Cevik, A. Zaidi, J. Rasheed, Lightweight CNN for accurate brain tumor detection from MRI with limited training data, Front. Med. 12 (2025) 1636059. https://doi.org/10.3389/fmed.2025.1636059

    work page doi:10.3389/fmed.2025.1636059 2025

  36. [36]

    Alotaibi, A

    M. Alotaibi, A. Aljouie, N. Alluhaidan, W. Qureshi, H. Almatar, R. Alduhayan, B. Alsomaie, A. Almazroa, Breast cancer classification b ased on convolutional neural network and image fusion approaches using ultrasound images, Heliyon 9 (2023) e22406. https://doi.org/10.1016/j.heliyon.2023.e22406

    work page doi:10.1016/j.heliyon.2023.e22406 2023

  37. [37]

    Ahmed, S

    H. Ahmed, S. Hamad, H.A. Shedeed, A.S. Hussein, Enhanced Deep Learning Model for Personalized Cancer Treatment, IEEE Access 10 (2022) 106050–106058. https://doi.org/10.1109/ACCESS.2022.3209285

    work page doi:10.1109/access.2022.3209285 2022

  38. [38]

    J. Gu, T. Tong, D. Xu, F. Cheng, C. Fang, C. He, J. Wang, B. Wang, X. Yang, K. Wang, J. Tian, T. Jiang, Deep learning radiomics of ultrasonography for comprehensively predicting tumor and axillary lymph node status after neoadjuvant chemotherapy in breast cancer patients: A multicenter study, Cancer 129 (2023) 356 –366. https://doi.org/10.1002/cncr.34540

    work page doi:10.1002/cncr.34540 2023

  39. [39]

    Yamashita, M

    R. Yamashita, M. Nishio, R.K.G. Do, K. Togashi, Convolutional neural networks: an overview and application in radiology, Insights Imaging 9 (2018) 611–629. https://doi.org/10.1007/s13244-018-0639-9

    work page doi:10.1007/s13244-018-0639-9 2018

  40. [40]

    Conceptual Understanding of Convolutional Neural Network – A Deep Learning Approach

    S. Indolia, A.K. Goswami, S.P. Mishra, P. Asopa, Conceptual Understanding of Convolutional Neural Network- A Deep Learning Approach, Procedia Computer Science 132 (2018) 679 –688. https://doi.org/10.1016/j.procs.2018.05.069

    work page doi:10.1016/j.procs.2018.05.069 2018

  41. [41]

    LeCun, Y

    Y. LeCun, Y. Bengio, G. Hinton, Deep learning, Nature 521 (2015) 436 –444. https://doi.org/10.1038/nature14539

    work page doi:10.1038/nature14539 2015

  42. [42]

    Taye, Theoretical Unde rstanding of Convolutional Neural Network: Concepts, Architectures, Applications, Future Directions, Computation 11 (2023) 52

    M.M. Taye, Theoretical Unde rstanding of Convolutional Neural Network: Concepts, Architectures, Applications, Future Directions, Computation 11 (2023) 52. https://doi.org/10.3390/computation11030052

    work page doi:10.3390/computation11030052 2023

  43. [43]

    Yogananda, B.R

    C.G.B. Yogananda, B.R. Shah, S.S. Nalawade, G.K. Murugesan, F.F. Yu, M.C. Pinho, B.C. Wagner, B. Mickey, T.R. Patel, B. Fei, A.J. Madhuranthakam, J.A. Maldjian, MRI -Based Deep-Learning Method for Determining Glioma MGMT Promoter Methylation Status, AJNR Am J Neuroradiol 42 (2021) 845 –852. https://doi.org/10.3174/ajnr.A7029

    work page doi:10.3174/ajnr.a7029 2021

  44. [44]

    Alyahya, A

    R. Alyahya, A. Alruwayqi, A. Alqarni, A. Alkhaldi, M. Alkubeyyer, X. Gao, M. Alshahrani, Multi -View MRI Approach for Classification of MGMT Methylation in Glioblastoma Patients, (2025). https://doi.org/10.48550/ARXIV.2512.14232

    work page doi:10.48550/arxiv.2512.14232 2025

  45. [45]

    Saeed, M

    N. Saeed, M. Ridzuan, H. Alasmawi, I. Sobirov, M. Yaqub, MGMT promoter methylation status prediction using MRI scans? An extensive experimental evaluation of deep learning models, Medical Image Analysis 90 (2023) 102989. https://doi.org/10.1016/j.media.2023.102989

    work page doi:10.1016/j.media.2023.102989 2023

  46. [46]

    X. Chen, M. Zeng, Y. Tong, T. Zhang, Y. Fu, H. Li, Z. Zhang, Z. Cheng, X. Xu, R. Yang, Z. Liu, X. Wei, X. Jiang, Automatic Prediction of MGMT Status in Glioblastoma via Deep Learning‐ Based MR Image Analysis, BioMed Research International 2020 (2020) 9258649. https://doi.org/10.1155/2020/9258649

    work page doi:10.1155/2020/9258649 2020

  47. [47]

    Faghani, B

    S. Faghani, B. Khosravi, M. Moassefi, G.M. Conte, B.J. Erickson, A Comparison of Three Different Deep Learning-Based Models to Predict the MGMT Promoter Methylation Status in Glioblastoma Using Brain MRI, J Digit Imaging 36 (2023) 837–846. https://doi.org/10.1007/s10278-022-00757-x

    work page doi:10.1007/s10278-022-00757-x 2023

  48. [48]

    Pálsson, S

    S. Pálsson, S. Cerri, K. Van Leemput, Prediction of MGMT Methylation Status of Glioblastoma Using Radiomics and Latent Space Shape Features, in: A. Crimi, S. Bakas (Eds.), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries, Springer International Publishing, Cham, 2022: pp. 222 –231. https://doi.org/10.1007/978-3-031-09002-8_20

    work page doi:10.1007/978-3-031-09002-8_20 2022

  49. [49]

    A. Ajay, R. Karthik, A.S. Bisht, A.K. Singh, QDeepColonNet: a quantum -based deep learning network for colorectal cancer classification using attention -driven DenseNet and shuffled dynamic local feature extraction network, Artif Intell Rev 58 (2025) 304. https://doi.org/10.1007/s10462-025-11295-7

    work page doi:10.1007/s10462-025-11295-7 2025

  50. [50]

    Hussein, A.M

    A.A. Hussein, A.M. Montaser, H.A. Elsayed, Skin cancer image classification using hybrid quantum deep learning model with BiLSTM and MobileNetV2, Quantum Mach. Intell. 7 (2025) 66. https://doi.org/10.1007/s42484-025-00288-y

    work page doi:10.1007/s42484-025-00288-y 2025

  51. [51]

    Idress, Y

    W.M. Idress, Y. Zhao, K.A. Abouda, H.M. Elhag, QCNN-Swin-UNet: Quantum Convolutional Neural Network Integrated with Optimized Swin -UNet for Efficient Liver Tumor Segmentation and Classification on Edge Devices, J Digit Imaging. Inform. Med. (2025). https://doi.org/10.1007/s10278-025-01630-3

    work page doi:10.1007/s10278-025-01630-3 2025

  52. [52]

    Ticku, V

    A. Ticku, V. Sangwan, S. Balani, S. Jha, S. Rawat, A. Rathee, D. Yadav, Advancing neuroimaging with quantum convolutional neural networks for brain tumor detection, Int. j. Inf. Tecnol. 17 (2025) 5759 –5766. https://doi.org/10.1007/s41870-025-02401-7

    work page doi:10.1007/s41870-025-02401-7 2025

  53. [53]

    Pandey, S

    P. Pandey, S. Mandal, A hybrid quantum –classical convolutional neural network with a quantum attention mechanism for skin cancer, Sci Rep 16 (2025) 1639. https://doi.org/10.1038/s41598-025-31122-x

    work page doi:10.1038/s41598-025-31122-x 2025

  54. [54]

    Short-Depth Circuits for Dicke State Preparation,

    E. Akpinar, N.M. Duc, B. KesercI, The Role of Quantum -enhanced Support Vector Machine using Multiparametric MRI Parameters in Differentiating Medulloblastoma from Ependymoma, in: 2022 IEEE International Conference on Quantum Computing and Engineering (QCE), IEEE, Broomfield, CO, USA, 2022: pp. 882–885. https://doi.org/10.1109/QCE53715.2022.00152

    work page doi:10.1109/qce53715.2022.00152 2022

  55. [55]

    E. Akpinar, Quantum Machine Learning in the Cognitive Domain: Alzheimer’s Disease Study, in: 2024 IEEE High Performance Extreme Computing Conference (HPEC), IEEE, Wakefield, MA, USA, 2024: pp. 1 –6. https://doi.org/10.1109/hpec62836.2024.10938482

    work page doi:10.1109/hpec62836.2024.10938482 2024

  56. [56]

    Akpinar, B

    E. Akpinar, B. Hangun, M. Oduncuoglu, O. Altun, O. Eyecioglu, Z. Yalcin, Quantum -Enhanced Classification of Brain Tumors Using DNA Microarray Gene Expression Profiles, in: 2025 IEEE Computer Society Annual Symposium on VLSI (ISVLSI), IEEE, Kalamata, Greece , 2025: pp. 1 –6. https://doi.org/10.1109/ISVLSI65124.2025.11130207

    work page doi:10.1109/isvlsi65124.2025.11130207 2025

  57. [57]

    Akpinar, S.M.N

    E. Akpinar, S.M.N. Islam, M. Oduncuoglu, Multi-classification of brain tumors using proposed hybrid quantum– classical integrated neural network (HQCINN) models: shallow and deep circuit approaches, Neural Comput & Applic 37 (2025) 22891–22922. https://doi.org/10.1007/s00521-025-11522-w

    work page doi:10.1007/s00521-025-11522-w 2025

  58. [58]

    Akpinar, M

    E. Akpinar, M. Oduncuoglu, Quantum Model Parallelism for MRI-Based Classification of Alzheimer’s Disease Stages, (2026). https://doi.org/10.48550/ARXIV.2602.00128

    work page doi:10.48550/arxiv.2602.00128 2026

  59. [59]

    Akpinar, M

    E. Akpinar, M. Oduncuoglu, Hybrid classical and quantum computing for enhanced glioma tumor classification using TCGA data, Sci Rep 15 (2025) 25935. https://doi.org/10.1038/s41598-025-97067-3

    work page doi:10.1038/s41598-025-97067-3 2025

  60. [60]

    In: 2024 IEEE Interna- tional Conference on Big Data (BigData), pp

    T. Tasnim, M. Rahman, F. Wu, Comparison of CNN and QCNN Performance in Binary Classification of Breast Cancer Histopathological Images, in: 2024 IEEE International Conference on Big Data (BigData), IEEE, Washington, DC, USA, 2024: pp. 3780–3787. https://doi.org/10.1109/BigData62323.2024.10825102

    work page doi:10.1109/bigdata62323.2024.10825102 2024

  61. [61]

    Liu, K.H

    J. Liu, K.H. Lim, K.L. Wood, W. Huang, C. Guo, H.-L. Huang, Hybrid quantum-classical convolutional neural networks, Sci. China Phys. Mech. Astron. 64 (2021) 290311. https://doi.org/10.1007/s11433-021-1734-3

    work page doi:10.1007/s11433-021-1734-3 2021

  62. [62]

    L.-H. Gong, J. -J. Pei, T. -F. Zhang, N. -R. Zhou, Quantu m convolutional neural network based on variational quantum circuits, Optics Communications 550 (2024) 129993. https://doi.org/10.1016/j.optcom.2023.129993

    work page doi:10.1016/j.optcom.2023.129993 2024

  63. [63]

    Li, R.-G

    Y. Li, R.-G. Zhou, R. Xu, J. Luo, W. Hu, A quantum deep convolutional neural network for image recognition, Quantum Sci. Technol. 5 (2020) 044003. https://doi.org/10.1088/2058-9565/ab9f93

    work page doi:10.1088/2058-9565/ab9f93 2020

  64. [64]

    Henderson, S

    M. Henderson, S. Shakya, S. Pradhan, T. Cook, Quanvolutional neural networks: powering image recognition with quantum circuits, Quantum Mach. Intell. 2 (2020) 2. https://doi.org/10.1007/s42484-020-00012-y

    work page doi:10.1007/s42484-020-00012-y 2020

  65. [65]

    Quantum machine learning in feature Hilbert spaces

    M. Schuld, N. Killoran, Quantum Machine Learning in Feature Hilbert Spaces, Phys. Rev. Lett. 122 (2019) 040504. https://doi.org/10.1103/PhysRevLett.122.040504

    work page doi:10.1103/physrevlett.122.040504 2019

  66. [66]

    Machine behaviour

    V. Havlíček, A.D. Córcoles, K. Temme, A.W. Harrow, A . Kandala, J.M. Chow, J.M. Gambetta, Supervised learning with quantum -enhanced feature spaces, Nature 567 (2019) 209 –212. https://doi.org/10.1038/s41586- 019-0980-2

    work page doi:10.1038/s41586- 2019

  67. [67]

    I. Cong, S. Choi, M.D. Lukin, Quantum convolutional neural networks, Nat. Phys. 15 (2019) 1273–1278. https://doi.org/10.1038/s41567-019-0648-8

    work page doi:10.1038/s41567-019-0648-8 2019

  68. [68]

    C. Long, M. Huang, X. Ye, Y. Futamura, T. Sakurai, Hybrid quantum -classical-quantum convolutional neural networks, Sci Rep 15 (2025) 31780. https://doi.org/10.1038/s41598-025-13417-1

    work page doi:10.1038/s41598-025-13417-1 2025

  69. [69]

    F. Fan, Y. Shi, T. Guggemos, X.X. Zhu, Hybrid Quantum -Classical Convolutional Neural Network Model for Image Classification, IEEE Trans. Neural Netw. Learning Syst. 35 (2024) 18145 –18159. https://doi.org/10.1109/TNNLS.2023.3312170

    work page doi:10.1109/tnnls.2023.3312170 2024

  70. [70]

    Flanders, C

    A. Flanders, C. Carr, E. Calabrese, F. Kitamura, J. Rudie, J. Mongan, J. Elliott, L. Prevedello, M. Riopel, S. Bakas, U. Baid, RSNA -MICCAI Brain Tumor Radiogenomic Classification, (2021). https://www.kaggle.com/competitions/rsna-miccai-brain-tumor-radiogenomic-classification

    2021

  71. [71]

    Mustafa, S

    W. Mustafa, S. Ali, N. Elgendy, S. Salama, L. El Sorogy, M. Mohsen, Role of contrast -enhanced FLAIR MRI in diagnosis of intracranial lesions, Egypt J Neurol Psychiatry Neurosurg 57 (2021) 108. https://doi.org/10.1186/s41983-021-00360-x

    work page doi:10.1186/s41983-021-00360-x 2021

  72. [72]

    Y. Yan, C. Yang, W. Chen, Z. Jia, H . Zhou, Z. Di, L. Xu, Multimodal MRI and artificial intelligence: Shaping the future of glioma, Journal of Neurorestoratology 13 (2025) 100175. https://doi.org/10.1016/j.jnrt.2024.100175

    work page doi:10.1016/j.jnrt.2024.100175 2025

  73. [73]

    Robinson, S

    S.D. Robinson, S. Kingdon, S.T. Williams, C.S. Hill, M. Williams, E. Chandy, G. Critchley, the Histo-Mol GBM collaborative, Understanding the difference in symptoms and outcomes between glioblastoma patients diagnosed based on histological or molecular criteria: a retrospective cohort analysis from the Histo -Mol GBM collaborative, J Neurooncol 176 (2026)...

    work page doi:10.1007/s11060-025-05364-8 2026

  74. [74]

    Han, M.R

    L. Han, M.R. Kamdar, MRI to MGMT: predicting methylation status in glioblastoma patients using convolutional recurrent neural networks, Pac Symp Biocomput 23 (2018) 331–342

    2018

  75. [75]

    Supervised Learning with Quantum Computers

    M. Schuld, F. Petruccione, Supervised Learning with Quantum Computers, Springer International Publishing, Cham, 2018. https://doi.org/10.1007/978-3-319-96424-9

    work page doi:10.1007/978-3-319-96424-9 2018

  76. [76]

    Nature549(7671), 195–202 (2017) https://doi.org/10.1038/nature23474

    J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, S. Lloyd, Quantum machine learning, Nature 549 (2017) 195–202. https://doi.org/10.1038/nature23474

    work page doi:10.1038/nature23474 2017

  77. [77]

    Schuld and F

    M. Schuld, F. Petruccione, Machine Learning with Quantum Computers, Springer International Publishing, Cham, 2021. https://doi.org/10.1007/978-3-030-83098-4

    work page doi:10.1007/978-3-030-83098-4 2021

  78. [78]

    Quantum embeddings for machine learning

    S. Lloyd, M. Schuld, A. I jaz, J. Izaac, N. Killoran, Quantum embeddings for machine learning, (2020). https://doi.org/10.48550/arXiv.2001.03622

    work page doi:10.48550/arxiv.2001.03622 2020

  79. [79]

    T. Hur, L. Kim, D.K. Park, Quantum convolutional neural network for classical data classification, Quantum Mach. Intell. 4 (2022) 3. https://doi.org/10.1007/s42484-021-00061-x

    work page doi:10.1007/s42484-021-00061-x 2022

  80. [80]

    Zeguendry, Z

    A. Zeguendry, Z. Jarir, M. Quafafou, Quantum Machine Learning: A Review and Case Studies, Entropy 25 (2023) 287. https://doi.org/10.3390/e25020287

    work page doi:10.3390/e25020287 2023

Showing first 80 references.