Cross modality medical image synthesis for improving liver segmentation

Deep learning-based computer-aided diagnosis (CAD) of medical images requires large datasets. However, the lack of large publicly available labelled datasets limits the development of deep learning-based CAD systems. Generative Adversarial Networks (GANs), in particular, CycleGAN, can be used to gen...

Full description

Saved in:
Bibliographic Details
Main Authors: Muhammad Rafiq, Hazrat Ali, Ghulam Mujtaba, Zubair Shah, Shoaib Azmat
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/21681163.2025.2476702
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850140005182210048
author Muhammad Rafiq
Hazrat Ali
Ghulam Mujtaba
Zubair Shah
Shoaib Azmat
author_facet Muhammad Rafiq
Hazrat Ali
Ghulam Mujtaba
Zubair Shah
Shoaib Azmat
author_sort Muhammad Rafiq
collection DOAJ
description Deep learning-based computer-aided diagnosis (CAD) of medical images requires large datasets. However, the lack of large publicly available labelled datasets limits the development of deep learning-based CAD systems. Generative Adversarial Networks (GANs), in particular, CycleGAN, can be used to generate new cross-domain images without paired training data. However, most CycleGAN-based synthesis methods lack the potential to overcome alignment and asymmetry between the input and generated data. We propose a two-stage technique for the synthesis of abdominal MRI using cross-modality translation of abdominal CT. We show that the synthetic data can help improve the performance of the liver segmentation network. We increase the number of abdominal MRI images through cross-modality image transformation of unpaired CT images using a CycleGAN inspired deformation invariant network called EssNet. Subsequently, we combine the synthetic MRI images with the original MRI images and use them to improve the accuracy of the U-Net on a liver segmentation task. We train the U-Net on real MRI images and then on real and synthetic MRI images. Consequently, by comparing both scenarios, we achieve an improvement in the performance of U-Net. In summary, the improvement achieved in the Intersection over Union (IoU) is 1.17%. The results show the potential to address the data scarcity challenge in medical imaging.
format Article
id doaj-art-59df717344574e8c8eb7cdc0a5064bf8
institution OA Journals
issn 2168-1163
2168-1171
language English
publishDate 2025-12-01
publisher Taylor & Francis Group
record_format Article
series Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization
spelling doaj-art-59df717344574e8c8eb7cdc0a5064bf82025-08-20T02:30:00ZengTaylor & Francis GroupComputer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization2168-11632168-11712025-12-0113110.1080/21681163.2025.2476702Cross modality medical image synthesis for improving liver segmentationMuhammad Rafiq0Hazrat Ali1Ghulam Mujtaba2Zubair Shah3Shoaib Azmat4Department of Electrical and Computer Engineering, COMSATS University Islamabad, Abbottabad, PakistanComputing Science and Mathematics, University of Stirling, Stirling, UKDepartment of Electrical and Computer Engineering, COMSATS University Islamabad, Abbottabad, PakistanCollege of Science and Engineering, Hamad Bin Khalifa University, Doha, QatarDepartment of Electrical and Computer Engineering, COMSATS University Islamabad, Abbottabad, PakistanDeep learning-based computer-aided diagnosis (CAD) of medical images requires large datasets. However, the lack of large publicly available labelled datasets limits the development of deep learning-based CAD systems. Generative Adversarial Networks (GANs), in particular, CycleGAN, can be used to generate new cross-domain images without paired training data. However, most CycleGAN-based synthesis methods lack the potential to overcome alignment and asymmetry between the input and generated data. We propose a two-stage technique for the synthesis of abdominal MRI using cross-modality translation of abdominal CT. We show that the synthetic data can help improve the performance of the liver segmentation network. We increase the number of abdominal MRI images through cross-modality image transformation of unpaired CT images using a CycleGAN inspired deformation invariant network called EssNet. Subsequently, we combine the synthetic MRI images with the original MRI images and use them to improve the accuracy of the U-Net on a liver segmentation task. We train the U-Net on real MRI images and then on real and synthetic MRI images. Consequently, by comparing both scenarios, we achieve an improvement in the performance of U-Net. In summary, the improvement achieved in the Intersection over Union (IoU) is 1.17%. The results show the potential to address the data scarcity challenge in medical imaging.https://www.tandfonline.com/doi/10.1080/21681163.2025.2476702Computer aided diagnosisCycleGANmedical imagingMRIsegmentation
spellingShingle Muhammad Rafiq
Hazrat Ali
Ghulam Mujtaba
Zubair Shah
Shoaib Azmat
Cross modality medical image synthesis for improving liver segmentation
Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization
Computer aided diagnosis
CycleGAN
medical imaging
MRI
segmentation
title Cross modality medical image synthesis for improving liver segmentation
title_full Cross modality medical image synthesis for improving liver segmentation
title_fullStr Cross modality medical image synthesis for improving liver segmentation
title_full_unstemmed Cross modality medical image synthesis for improving liver segmentation
title_short Cross modality medical image synthesis for improving liver segmentation
title_sort cross modality medical image synthesis for improving liver segmentation
topic Computer aided diagnosis
CycleGAN
medical imaging
MRI
segmentation
url https://www.tandfonline.com/doi/10.1080/21681163.2025.2476702
work_keys_str_mv AT muhammadrafiq crossmodalitymedicalimagesynthesisforimprovingliversegmentation
AT hazratali crossmodalitymedicalimagesynthesisforimprovingliversegmentation
AT ghulammujtaba crossmodalitymedicalimagesynthesisforimprovingliversegmentation
AT zubairshah crossmodalitymedicalimagesynthesisforimprovingliversegmentation
AT shoaibazmat crossmodalitymedicalimagesynthesisforimprovingliversegmentation