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

International Journal of Computer Assisted Radiology and Surgery

Vision transformer embeddings and quantum pyramidal circuits for biomedical image analysis

Xavier Font-Aragonès, Miguel Á. González Ballester

Published May 25, 2026• 0 citations
Publisher

Abstract

Abstract This work presents a novel quantum-hybrid pipeline for lung nodule classification in computed tomography (CT) scans, combining vision transformer (ViT) embeddings with quantum orthogonal pyramidal circuits (QOPCs). The approach was evaluated on 681 lung nodule CT scans across axial, coronal, and sagittal planes. Two ViT configurations were tested: ViT 1 (1 head, 4 layers) and a Bayesian-optimized ViT 2 (4 heads, 8 layers). Features extracted from ViT embedding layers were reduced via principal component analysis to 2–16 dimensions and classified using the QOPC with reconfigurable beam splitter (RBS) gates. The proposed approach achieved unprecedented compression, up to 1,470 $$\times $$ × (from 10,290 to 7 parameters) while preserving over 99% of baseline accuracy. The approach reached 83.7% accuracy (ViT 2 , $$k{=}8$$ k = 8 , $$h{=}8$$ h = 8 ) with only 46 trainable parameters and achieved computational efficiency (CE) up to 92.0. Training was accelerated up to 28 $$\times $$ × (0.030 vs. 0.833 min) while maintaining robust diagnostic performance (F1: 0.77−0.82, receiver operating characteristic area under the curve (ROC-AUC): 0.87−0.90). Ablation studies confirmed that the quantum layer outperforms conventional MLPs by +3.4% accuracy with 35% fewer parameters, while late fusion of multi-view predictions further improved performance to 85.4% accuracy and 0.92 ROC-AUC. These results establish hybrid ViT-QOPC architectures as a practical and resource-efficient framework for medical image analysis, demonstrating their ability to dramatically reduce computational cost without compromising clinical accuracy.

DOI
10.1007/s11548-026-03696-y

Related papers

A novel noise-resilient quantum-classical vision transformer with quantum state tomography for high-precision lung cancer classification

Sumaya Mustofa, Yousuf Rayhan Emon, Md․Zahid Hasan +1 more

Lung cancer remains the leading cause of cancer-related mortality worldwide, with accurate subtype classification being critical for improving patient outcomes. Vision Transformers (ViTs) have recently advanced medical image analysis, but their integration with quantum computing remains underexplored, leaving a gap in leveraging quantum advantages for clinical applications. This study proposes a quantum-enhanced Vision Transformer (QViT) that embeds variational quantum circuits (VQCs) within attention and feed-forward layers to enrich feature learning. To assess robustness, two circuit families, Basic VQC and Quantum Approximate Optimization Algorithm (QAOA) based VQC, were evaluated across two depths, yielding four QViT configurations. The models were analyzed through cancer classification tasks, systematic quantum state tomography (QST) to examine quantum state properties, and noise resilience testing under realistic Noisy Intermediate-Scale Quantum (NISQ) conditions, reflecting current hardware limitations. On 3150 CT (Computed Tomography) images, the QViT-QAOA-D1 configuration achieved 98.52% accuracy with a training time of 674 s, outperforming other configurations. Quantum analyses further showed that Basic VQC provides stronger entanglement but requires deeper, less NISQ-friendly circuits, whereas QAOA maintains high-purity states at shallower depths. These results suggest that QViT provides a practical and scalable framework, where quantum-enhanced representations can improve diagnostic performance while remaining compatible with current hardware constraints. Future work will focus on validation using patient-level datasets and deployment on real quantum hardware to further assess clinical applicability and scalability.

Computers & Electrical Engineering•Jun 8, 2026•0 citations

MedQViT: A Lightweight Hybrid Quantum Vision Transformer for Medical Image Analysis

Amandeep Singh Bhatia, Mandeep Kaur Saggi

Classical Vision Transformers (CViTs) have demonstrated strong performance in medical image analysis, supporting a wide range of digital health applications. However, their self-attention mechanisms can require substantial computational resources and model parameters. In parallel, quantum machine learning (QML) has attracted increasing interest in healthcare and other scientific domains, particularly because quantum-enhanced models may provide expressive representations for small datasets. In this paper, we propose MedQViT, a robust and parameter-efficient Quantum Vision Transformer for medical image classification. To reduce the computational cost of global self-attention, MedQViT combines a shared four-qubit parameterized quantum circuit with a quantum-feature state memory and trainable landmark tokens. This hybrid architecture encodes image patches into compact quantum-enhanced representations while using classical layers for feature aggregation and classification. Experiments on established medical image datasets evaluate the effectiveness and feasibility of MedQViT against classical and quantum baselines. MedQViT achieves competitive performance with substantially fewer parameters than conventional vision models and outperforms several classical and quantum methods in terms of test accuracy (ACC) and area under the receiver operating characteristic curve (AUC).

Proceedings of the AAAI Symposium Series•Nov 6, 2026•0 citations

Hybrid Quantum Vision Transformer for Lung Cancer Detection from CT Scans

Bissen Abbassi, Amor Gueddana, Haythem Ghazouani +2 more

Early detection of lung cancer via computed tomography (CT) scans is critical for clinical outcomes, yet the computational cost of state-of-the-art Vision Transformers limits their applicability to high-resolution volumetric data. We propose the Recurrent Quantum Self-Attention Network (RQSAN), a hybrid model combining a gated recurrent unit (GRU) contextual encoder with a measurement-free quantum self-attention module, designed to operate within the qubit and circuit-depth limits of current quantum simulators. The original QSAN evaluation [5] is restricted to the binary MNIST 0-vs-1 task at 4 × 4 resolution and includes no classical baselines. We extend this evaluation to, first, harder MNIST pairs at 8 × 8, second, the LIDC-IDRI lung nodule classification task at the same resolution, and third, matched-resolution classical baselines (logistic regression, parameter-matched MLPs).On 0-vs-1, RQSAN reproduces the published near-ceiling performance with accuracy 0.995. On the harder 4-vs-9 pair, RQSAN’s ranking ability matches the parameter-matched MLP at AUROC 0.907, while threshold accuracy collapses, revealing a readout-bandwidth bottleneck. On LIDC-IDRI at 8 × 8, all model families reach only near-chance AUROC of 0.45–0.49, revealing a second bottleneck in input resolution. We provide a fair evaluation protocol and identify the higher-resolution input-encoding problem as the central challenge for quantum-enhanced medical imaging.

Procedia Computer Science•Jan 1, 2026•0 citations

Quantum-Classical Synergy in Medical AI: A Dual Phase Benchmark for Diagnostic Intelligence in Lung Cancer Classification

Muhammad Bilal Khan, Saqib Nazir

Abstract Pulmonary nodule classification remains constrained by variability in computed tomography (CT) acquisition protocols, annotation quality, and the extreme class imbalance that hinders accurate decision–making in diagnosis. Traditional deep learning approaches have been severely limited by the lack of diversity in training datasets, their tendency to be overly sensitive to noise, and their inability to account for the large number of nonlinear feature interactions present in volumetric structures. Therefore, a dual–phase framework is presented to investigate the application of classical and hybrid quantum–classical (HQC) architectures. Phase 1 evaluates CNN and hybrid Quantum–classical (HQC) feature extractors to quantify quantum expressivity. The HQC model employs a series of layered variational quantum circuits (VQCs) with calibrated rotation gates and structured entanglement patterns to expand the hypothesis space while improving operational stability. HQC achieved superior ROC–AUC (0.89 ± 0.02) and recall (0.86 ± 0.03). Phase 2 benchmarks the performance of Swin–Transformer embeddings with classical classifiers, as well as the HQC for a fair comparison. The HQC constantly outperforms other methods (ROC–AUC = 0.73 ± 0.02), indicating robustly enhanced quantum representations. Conditional Generative Adversarial Networks (CGANs) were used to address the extreme dataset imbalance. This dissertation provides statistically valid evidence that HQC architectures significantly increase both classification sensitivity and stability, thereby offering an opportunity for future research on precision oncology under high-variance, data-constrained conditions.

•Jun 22, 2026•0 citations

QResNet: A Hybrid Quantum-Classical Network For Medical Image Classification

Dongxue Bai, Qi Li, Dongfen Li +3 more

Medical image classification has been a fundamental task in computer-aided diagnosis. In this work, we propose a hybrid quantum-classical deep learning framework that integrates a parameterized quantum layer into the ResNet-18 architecture for medical image classification. Our experiments are conducted on the OrganMNIST-Axial, OrganMNIST-Coronal, and OrganMNIST-Sagittal datasets from the MedMNIST v2 collection, each consisting of $28 \times 28$ grayscale images representing different organ sections. The quantum layer is implemented as a 6-qubit variational quantum circuit (VQC) that partially replaces the classical fully connected module, enabling the network to exploit quantum feature mapping and entanglement for enhanced nonlinear representation learning. This architectural modification reduces the number of classical trainable parameters by approximately 44% compared with the original ResNet-18 while preserving its representational capacity. We evaluate the proposed QResNet18 against classical baselines to investigate its potential advantages in terms of accuracy, robustness, and parameter efficiency. Experimental results demonstrate that the hybrid quantum-classical model achieves competitive or superior performance while maintaining a significantly reduced parameter count, highlighting the feasibility and promise of quantumenhanced medical image classification.

International Conference on Cloud Computing, Big Data Application and Software Engineering (CBASE)•Oct 1, 2025•1 citation

Quantum enhanced machine learning for medical image analysis: a hybrid approach

Soha Rawas, Duaa AlSaeed

Purpose This study investigates the integration of quantum-enhanced machine learning (QML) into medical image analysis, focusing on classification and feature extraction tasks. It aims to evaluate whether hybrid quantum-classical models can overcome the limitations of traditional machine learning in handling high-dimensional, clinically relevant imaging data. Design/methodology/approach Hybrid architectures combining classical preprocessing with quantum components, specifically variational quantum circuits (VQCs) and quantum kernel methods, were developed and implemented in simulated quantum environments using Qiskit and PennyLane on Google Colab. Publicly available datasets (ChestX-ray14, Brain Tumor MRI and CT Hemorrhage) were preprocessed and encoded into quantum states. Model performance was evaluated against classical baselines (MLP and CNN) using accuracy, precision, recall, F1-score, convergence speed and trainable parameters. Findings Quantum-enhanced models achieved competitive or superior performance compared to classical models with significantly fewer trainable parameters and faster convergence. The Quantum Kernel SVM demonstrated the highest precision (89.1%) and F1-score (88.5%) using only 8 parameters, while VQC models converged up to 2× faster than classical baselines. These results highlight the potential of QML to improve efficiency, robustness and scalability in diagnostic imaging workflows, particularly in resource-constrained or data-limited settings. Research limitations/implications The study relies on simulated quantum environments, which do not capture hardware noise or qubit limitations. Future work will extend to real quantum devices, multimodal data integration and interpretability strategies. Practical implications Quantum kernel methods and lightweight VQC architectures enable faster, resource-efficient diagnostic modeling, making them suitable for clinical environments with limited computational capacity or time-sensitive decision-making needs. Social implications Improving the efficiency and accessibility of medical image analysis has the potential to support faster diagnosis and enhance healthcare delivery, particularly in underserved or resource-limited regions. Originality/value This study provides a practical evaluation of hybrid quantum-classical models applied to multiple real medical imaging datasets in a reproducible simulated environment. It offers an early-stage framework demonstrating how QML techniques can be integrated into existing machine learning pipelines for medical image analysis.

Applied Computing and Informatics•Jan 27, 2026•5 citations
Abstract This work presents a novel quantum-hybrid pipeline for lung nodule classification in computed tomography (CT) scans, combining vision transformer (ViT) embeddings with quantum orthogonal pyramidal circuits (QOPCs). The approach was evaluated on 681 lung nodule CT scans across axial, coronal, and sagittal planes. Two ViT configurations were tested: ViT 1 (1 head, 4 layers) and a Bayesian-optimized ViT 2 (4 heads, 8 layers). Features extracted from ViT embedding layers were reduced via principal component analysis to 2–16 dimensions and classified using the QOPC with reconfigurable beam splitter (RBS) gates. The proposed approach achieved unprecedented compression, up to 1,470 ×\times × × (from 10,290 to 7 parameters) while preserving over 99% of baseline accuracy. The approach reached 83.7% accuracy (ViT 2 , k=8k{=}8k=8 k = 8 , h=8h{=}8h=8 h = 8 ) with only 46 trainable parameters and achieved computational efficiency (CE) up to 92.0. Training was accelerated up to 28 ×\times × × (0.030 vs. 0.833 min) while maintaining robust diagnostic performance (F1: 0.77−0.82, receiver operating characteristic area under the curve (ROC-AUC): 0.87−0.90). Ablation studies confirmed that the quantum layer outperforms conventional MLPs by +3.4% accuracy with 35% fewer parameters, while late fusion of multi-view predictions further improved performance to 85.4% accuracy and 0.92 ROC-AUC. These results establish hybrid ViT-QOPC architectures as a practical and resource-efficient framework for medical image analysis, demonstrating their ability to dramatically reduce computational cost without compromising clinical accuracy.
Lung cancer remains the leading cause of cancer-related mortality worldwide, with accurate subtype classification being critical for improving patient outcomes. Vision Transformers (ViTs) have recently advanced medical image analysis, but their integration with quantum computing remains underexplored, leaving a gap in leveraging quantum advantages for clinical applications. This study proposes a quantum-enhanced Vision Transformer (QViT) that embeds variational quantum circuits (VQCs) within attention and feed-forward layers to enrich feature learning. To assess robustness, two circuit families, Basic VQC and Quantum Approximate Optimization Algorithm (QAOA) based VQC, were evaluated across two depths, yielding four QViT configurations. The models were analyzed through cancer classification tasks, systematic quantum state tomography (QST) to examine quantum state properties, and noise resilience testing under realistic Noisy Intermediate-Scale Quantum (NISQ) conditions, reflecting current hardware limitations. On 3150 CT (Computed Tomography) images, the QViT-QAOA-D1 configuration achieved 98.52% accuracy with a training time of 674 s, outperforming other configurations. Quantum analyses further showed that Basic VQC provides stronger entanglement but requires deeper, less NISQ-friendly circuits, whereas QAOA maintains high-purity states at shallower depths. These results suggest that QViT provides a practical and scalable framework, where quantum-enhanced representations can improve diagnostic performance while remaining compatible with current hardware constraints. Future work will focus on validation using patient-level datasets and deployment on real quantum hardware to further assess clinical applicability and scalability.
Classical Vision Transformers (CViTs) have demonstrated strong performance in medical image analysis, supporting a wide range of digital health applications. However, their self-attention mechanisms can require substantial computational resources and model parameters. In parallel, quantum machine learning (QML) has attracted increasing interest in healthcare and other scientific domains, particularly because quantum-enhanced models may provide expressive representations for small datasets. In this paper, we propose MedQViT, a robust and parameter-efficient Quantum Vision Transformer for medical image classification. To reduce the computational cost of global self-attention, MedQViT combines a shared four-qubit parameterized quantum circuit with a quantum-feature state memory and trainable landmark tokens. This hybrid architecture encodes image patches into compact quantum-enhanced representations while using classical layers for feature aggregation and classification. Experiments on established medical image datasets evaluate the effectiveness and feasibility of MedQViT against classical and quantum baselines. MedQViT achieves competitive performance with substantially fewer parameters than conventional vision models and outperforms several classical and quantum methods in terms of test accuracy (ACC) and area under the receiver operating characteristic curve (AUC).
Early detection of lung cancer via computed tomography (CT) scans is critical for clinical outcomes, yet the computational cost of state-of-the-art Vision Transformers limits their applicability to high-resolution volumetric data. We propose the Recurrent Quantum Self-Attention Network (RQSAN), a hybrid model combining a gated recurrent unit (GRU) contextual encoder with a measurement-free quantum self-attention module, designed to operate within the qubit and circuit-depth limits of current quantum simulators. The original QSAN evaluation [5] is restricted to the binary MNIST 0-vs-1 task at 4 × 4 resolution and includes no classical baselines. We extend this evaluation to, first, harder MNIST pairs at 8 × 8, second, the LIDC-IDRI lung nodule classification task at the same resolution, and third, matched-resolution classical baselines (logistic regression, parameter-matched MLPs).On 0-vs-1, RQSAN reproduces the published near-ceiling performance with accuracy 0.995. On the harder 4-vs-9 pair, RQSAN’s ranking ability matches the parameter-matched MLP at AUROC 0.907, while threshold accuracy collapses, revealing a readout-bandwidth bottleneck. On LIDC-IDRI at 8 × 8, all model families reach only near-chance AUROC of 0.45–0.49, revealing a second bottleneck in input resolution. We provide a fair evaluation protocol and identify the higher-resolution input-encoding problem as the central challenge for quantum-enhanced medical imaging.
Abstract Pulmonary nodule classification remains constrained by variability in computed tomography (CT) acquisition protocols, annotation quality, and the extreme class imbalance that hinders accurate decision–making in diagnosis. Traditional deep learning approaches have been severely limited by the lack of diversity in training datasets, their tendency to be overly sensitive to noise, and their inability to account for the large number of nonlinear feature interactions present in volumetric structures. Therefore, a dual–phase framework is presented to investigate the application of classical and hybrid quantum–classical (HQC) architectures. Phase 1 evaluates CNN and hybrid Quantum–classical (HQC) feature extractors to quantify quantum expressivity. The HQC model employs a series of layered variational quantum circuits (VQCs) with calibrated rotation gates and structured entanglement patterns to expand the hypothesis space while improving operational stability. HQC achieved superior ROC–AUC (0.89 ± 0.02) and recall (0.86 ± 0.03). Phase 2 benchmarks the performance of Swin–Transformer embeddings with classical classifiers, as well as the HQC for a fair comparison. The HQC constantly outperforms other methods (ROC–AUC = 0.73 ± 0.02), indicating robustly enhanced quantum representations. Conditional Generative Adversarial Networks (CGANs) were used to address the extreme dataset imbalance. This dissertation provides statistically valid evidence that HQC architectures significantly increase both classification sensitivity and stability, thereby offering an opportunity for future research on precision oncology under high-variance, data-constrained conditions.
Medical image classification has been a fundamental task in computer-aided diagnosis. In this work, we propose a hybrid quantum-classical deep learning framework that integrates a parameterized quantum layer into the ResNet-18 architecture for medical image classification. Our experiments are conducted on the OrganMNIST-Axial, OrganMNIST-Coronal, and OrganMNIST-Sagittal datasets from the MedMNIST v2 collection, each consisting of 28×2828 \times 2828×28 grayscale images representing different organ sections. The quantum layer is implemented as a 6-qubit variational quantum circuit (VQC) that partially replaces the classical fully connected module, enabling the network to exploit quantum feature mapping and entanglement for enhanced nonlinear representation learning. This architectural modification reduces the number of classical trainable parameters by approximately 44% compared with the original ResNet-18 while preserving its representational capacity. We evaluate the proposed QResNet18 against classical baselines to investigate its potential advantages in terms of accuracy, robustness, and parameter efficiency. Experimental results demonstrate that the hybrid quantum-classical model achieves competitive or superior performance while maintaining a significantly reduced parameter count, highlighting the feasibility and promise of quantumenhanced medical image classification.
Purpose This study investigates the integration of quantum-enhanced machine learning (QML) into medical image analysis, focusing on classification and feature extraction tasks. It aims to evaluate whether hybrid quantum-classical models can overcome the limitations of traditional machine learning in handling high-dimensional, clinically relevant imaging data. Design/methodology/approach Hybrid architectures combining classical preprocessing with quantum components, specifically variational quantum circuits (VQCs) and quantum kernel methods, were developed and implemented in simulated quantum environments using Qiskit and PennyLane on Google Colab. Publicly available datasets (ChestX-ray14, Brain Tumor MRI and CT Hemorrhage) were preprocessed and encoded into quantum states. Model performance was evaluated against classical baselines (MLP and CNN) using accuracy, precision, recall, F1-score, convergence speed and trainable parameters. Findings Quantum-enhanced models achieved competitive or superior performance compared to classical models with significantly fewer trainable parameters and faster convergence. The Quantum Kernel SVM demonstrated the highest precision (89.1%) and F1-score (88.5%) using only 8 parameters, while VQC models converged up to 2× faster than classical baselines. These results highlight the potential of QML to improve efficiency, robustness and scalability in diagnostic imaging workflows, particularly in resource-constrained or data-limited settings. Research limitations/implications The study relies on simulated quantum environments, which do not capture hardware noise or qubit limitations. Future work will extend to real quantum devices, multimodal data integration and interpretability strategies. Practical implications Quantum kernel methods and lightweight VQC architectures enable faster, resource-efficient diagnostic modeling, making them suitable for clinical environments with limited computational capacity or time-sensitive decision-making needs. Social implications Improving the efficiency and accessibility of medical image analysis has the potential to support faster diagnosis and enhance healthcare delivery, particularly in underserved or resource-limited regions. Originality/value This study provides a practical evaluation of hybrid quantum-classical models applied to multiple real medical imaging datasets in a reproducible simulated environment. It offers an early-stage framework demonstrating how QML techniques can be integrated into existing machine learning pipelines for medical image analysis.