Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

A hybrid quantum-classical framework for MRI-based deep brain tumor segmentation and classification.

Naglaa F Soliman1, Prashant Kumar Shukla2, Mohamed M Hassan3

  • 1Department of Information Technology, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University, P.O. Box 84428, Riyadh, 11671, Saudi Arabia.

Scientific Reports
|July 3, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Thyroid disease detection using enhanced extreme learning machine based on drop-connect method.

Scientific reports·2026
Same author

DFT-driven insights into (Sr/Ba)<sub>2</sub>GaBiO<sub>6</sub> double perovskites for next-generation optoelectronic and thermoelectric technologies.

Scientific reports·2026
Same author

Bridging modalities: a deep learning framework for brain tumor classification via CT-MRI integration and model fusion.

Frontiers in computational neuroscience·2026
Same author

Machine learning and response surface methodology for optimization and prediction of tribological performance of PLA/rice husk biochar composites.

Scientific reports·2026
Same author

Sixty-port dual-band octa-pentacle MIMO antenna for vehicular communications.

Scientific reports·2026
Same author

Efficient EOG-based movement classification in IoMT using machine learning algorithms for people with motor disabilities.

Disability and rehabilitation. Assistive technology·2026

This study introduces QFormer-Brain, a hybrid quantum-classical model for brain tumor segmentation and classification using MRI. The novel framework significantly improves diagnostic accuracy by leveraging quantum representation learning with transformers.

Area of Science:

  • Artificial Intelligence
  • Quantum Computing
  • Medical Imaging

Background:

  • Brain tumor characterization from MRI is vital but challenging due to tumor heterogeneity and complex imaging features.
  • Existing deep learning models struggle with intricate feature interactions and subtle pathological variations in MRI data.
  • Quantum machine learning offers potential for more expressive data representation through quantum feature encoding.

Purpose of the Study:

  • To propose and evaluate QFormer-Brain, a hybrid quantum-classical framework for automated brain tumor segmentation and classification.
  • To enhance feature representation by integrating quantum representation learning with transformer-based deep learning architectures.

Main Methods:

  • Utilized a Shifted Window UNet Transformers (Swin-UNETR) for initial tumor segmentation, capturing local and global contextual information.
Keywords:
Brain tumor segmentationMRI classificationQuantum machine learningQuantum transformer classifierSwin-UNETRVariational quantum circuit

Related Experiment Videos

  • Employed quantum feature mapping with angle encoding and variational quantum circuits (VQCs) to generate richer nonlinear feature representations.
  • Integrated quantum embeddings with transformer features for classification using a Quantum Transformer Classifier on the BraTS 2021 dataset.

    • Main Results:

    • Achieved high performance metrics: Dice Similarity Coefficient (DSC) of 97.1%, Intersection-over-Union (IoU) of 95.0%, and classification accuracy of 98.5%.
    • Demonstrated superior performance compared to established models like U-Net, nnU-Net, ResNet50, and Vision Transformer (ViT).
    • Validated the effectiveness of quantum-inspired representation learning in boosting feature extraction from complex MRI data.

    Conclusions:

    • The QFormer-Brain framework effectively combines transformer architectures with quantum representation learning for advanced brain tumor analysis.
    • Quantum-inspired methods offer a promising avenue for improving the discriminative power of deep learning models in medical imaging.
    • This hybrid approach shows significant potential for enhancing automated diagnosis and treatment planning for brain tumors.