Application of Quantum Convolutional Neural Networks for MRI-Based Brain Tumor Detection and Classification
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911134713380864 |
|---|---|
| author | Nugraha, Sugih Pratama Alfajri, Ariiq Islam Sumaryada, Tony Tai, Duong Thanh Tamam, Nissren Sulieman, Abdelmoneim Yani, Sitti |
| author_facet | Nugraha, Sugih Pratama Alfajri, Ariiq Islam Sumaryada, Tony Tai, Duong Thanh Tamam, Nissren Sulieman, Abdelmoneim Yani, Sitti |
| contents | This study explores the application of Quantum Convolutional Neural Networks (QCNNs) for brain tumor classification using MRI images, leveraging quantum computing for enhanced computational efficiency. A dataset of 3,264 MRI images, including glioma, meningioma, pituitary tumors, and non-tumor cases, was utilized. The data was split into 80% training and 20% testing, with an oversampling technique applied to address class imbalance. The QCNN model consists of quantum convolution layers, flatten layers, and dense layers, with a filter size of 2, depth of 4, and 4 qubits, trained over 10 epochs. Two models were developed: a binary classification model distinguishing tumor presence and a multiclass classification model categorizing tumor types. The binary model achieved 88% accuracy, improving to 89% after data balancing, while the multiclass model achieved 52% accuracy, increasing to 62% after oversampling. Despite strong binary classification performance, the multiclass model faced challenges due to dataset complexity and quantum circuit limitations. These findings suggest that QCNNs hold promise for medical imaging applications, particularly in binary classification. However, further refinements, including optimized quantum circuit architectures and hybrid classical-quantum approaches, are necessary to enhance multiclass classification accuracy and improve QCNN applicability in clinical settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_02582 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Application of Quantum Convolutional Neural Networks for MRI-Based Brain Tumor Detection and Classification Nugraha, Sugih Pratama Alfajri, Ariiq Islam Sumaryada, Tony Tai, Duong Thanh Tamam, Nissren Sulieman, Abdelmoneim Yani, Sitti Medical Physics Computer Vision and Pattern Recognition This study explores the application of Quantum Convolutional Neural Networks (QCNNs) for brain tumor classification using MRI images, leveraging quantum computing for enhanced computational efficiency. A dataset of 3,264 MRI images, including glioma, meningioma, pituitary tumors, and non-tumor cases, was utilized. The data was split into 80% training and 20% testing, with an oversampling technique applied to address class imbalance. The QCNN model consists of quantum convolution layers, flatten layers, and dense layers, with a filter size of 2, depth of 4, and 4 qubits, trained over 10 epochs. Two models were developed: a binary classification model distinguishing tumor presence and a multiclass classification model categorizing tumor types. The binary model achieved 88% accuracy, improving to 89% after data balancing, while the multiclass model achieved 52% accuracy, increasing to 62% after oversampling. Despite strong binary classification performance, the multiclass model faced challenges due to dataset complexity and quantum circuit limitations. These findings suggest that QCNNs hold promise for medical imaging applications, particularly in binary classification. However, further refinements, including optimized quantum circuit architectures and hybrid classical-quantum approaches, are necessary to enhance multiclass classification accuracy and improve QCNN applicability in clinical settings. |
| title | Application of Quantum Convolutional Neural Networks for MRI-Based Brain Tumor Detection and Classification |
| topic | Medical Physics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.02582 |