Automatic segmentation of white matter lesions on multi-parametric MRI: convolutional neural network versus vision transformer
Abstract Background and purpose White matter hyperintensities in brain MRI are key indicators of various neurological conditions, and their accurate segmentation is essential for assessing disease progression. This study aims to evaluate the performance of a 3D convolutional neural network and a 3D...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-01-01
|
| Series: | BMC Neurology |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12883-024-04010-6 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559364110909440 |
|---|---|
| author | Yun-Ting Chen Yan-Cheng Huang Hsiu-Ling Chen Hsin-Chih Lo Pei-Chin Chen Chiun-Chieh Yu Yi-Chin Tu Tyng-Luh Liu Wei-Che Lin |
| author_facet | Yun-Ting Chen Yan-Cheng Huang Hsiu-Ling Chen Hsin-Chih Lo Pei-Chin Chen Chiun-Chieh Yu Yi-Chin Tu Tyng-Luh Liu Wei-Che Lin |
| author_sort | Yun-Ting Chen |
| collection | DOAJ |
| description | Abstract Background and purpose White matter hyperintensities in brain MRI are key indicators of various neurological conditions, and their accurate segmentation is essential for assessing disease progression. This study aims to evaluate the performance of a 3D convolutional neural network and a 3D Transformer-based model for white matter hyperintensities segmentation, focusing on their efficacy with limited datasets and similar computational resources. Materials and methods We implemented a convolution-based model (3D ResNet-50 U-Net with spatial and channel squeeze & excitation) and a Transformer-based model (3D Swin Transformer with a convolutional stem). The models were evaluated on two clinical datasets from Kaohsiung Chang Gung Memorial Hospital and National Center for High-Performance Computing. Four metrics were used for evaluation: Dice similarity coefficient, lesion segmentation, lesion F1-Score, and lesion sensitivity. Results The Transformer-based model, with appropriate adjustments, outperformed the well-established convolution-based model in foreground Dice similarity coefficient, lesion F1-Score, and sensitivity, demonstrating robust segmentation accuracy. DRLoc enhanced the Transformer’s performance, achieving comparable results on internal and benchmark datasets despite limited data availability. Conclusion With comparable computational overhead, a Transformer-based model can surpass a well-established convolution-based model in white matter hyperintensities segmentation on small datasets by capturing global context effectively, making them suitable for clinical applications where computational resources are constrained. |
| format | Article |
| id | doaj-art-66138e55a9654852a06e9677d56ac543 |
| institution | Kabale University |
| issn | 1471-2377 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Neurology |
| spelling | doaj-art-66138e55a9654852a06e9677d56ac5432025-01-05T12:34:04ZengBMCBMC Neurology1471-23772025-01-0125111210.1186/s12883-024-04010-6Automatic segmentation of white matter lesions on multi-parametric MRI: convolutional neural network versus vision transformerYun-Ting Chen0Yan-Cheng Huang1Hsiu-Ling Chen2Hsin-Chih Lo3Pei-Chin Chen4Chiun-Chieh Yu5Yi-Chin Tu6Tyng-Luh Liu7Wei-Che Lin8Department of Diagnostic Radiology, Kaohsiung Chang Gung Memorial Hospital, Chang Gung University College of MedicineTaiwan AI LabsDepartment of Diagnostic Radiology, Kaohsiung Chang Gung Memorial Hospital, Chang Gung University College of MedicineTaiwan AI LabsDepartment of Diagnostic Radiology, Kaohsiung Chang Gung Memorial Hospital, Chang Gung University College of MedicineDepartment of Diagnostic Radiology, Kaohsiung Chang Gung Memorial Hospital, Chang Gung University College of MedicineTaiwan AI LabsTaiwan AI LabsDepartment of Diagnostic Radiology, Kaohsiung Chang Gung Memorial Hospital and Chang Gung University College of Medicine, School of Medicine, College of Medicine, National Sun Yat-Sen UniversityAbstract Background and purpose White matter hyperintensities in brain MRI are key indicators of various neurological conditions, and their accurate segmentation is essential for assessing disease progression. This study aims to evaluate the performance of a 3D convolutional neural network and a 3D Transformer-based model for white matter hyperintensities segmentation, focusing on their efficacy with limited datasets and similar computational resources. Materials and methods We implemented a convolution-based model (3D ResNet-50 U-Net with spatial and channel squeeze & excitation) and a Transformer-based model (3D Swin Transformer with a convolutional stem). The models were evaluated on two clinical datasets from Kaohsiung Chang Gung Memorial Hospital and National Center for High-Performance Computing. Four metrics were used for evaluation: Dice similarity coefficient, lesion segmentation, lesion F1-Score, and lesion sensitivity. Results The Transformer-based model, with appropriate adjustments, outperformed the well-established convolution-based model in foreground Dice similarity coefficient, lesion F1-Score, and sensitivity, demonstrating robust segmentation accuracy. DRLoc enhanced the Transformer’s performance, achieving comparable results on internal and benchmark datasets despite limited data availability. Conclusion With comparable computational overhead, a Transformer-based model can surpass a well-established convolution-based model in white matter hyperintensities segmentation on small datasets by capturing global context effectively, making them suitable for clinical applications where computational resources are constrained.https://doi.org/10.1186/s12883-024-04010-6White matter hyperintensitiesSegmentationBrain MRIConvolutional neural networkVision transformer |
| spellingShingle | Yun-Ting Chen Yan-Cheng Huang Hsiu-Ling Chen Hsin-Chih Lo Pei-Chin Chen Chiun-Chieh Yu Yi-Chin Tu Tyng-Luh Liu Wei-Che Lin Automatic segmentation of white matter lesions on multi-parametric MRI: convolutional neural network versus vision transformer BMC Neurology White matter hyperintensities Segmentation Brain MRI Convolutional neural network Vision transformer |
| title | Automatic segmentation of white matter lesions on multi-parametric MRI: convolutional neural network versus vision transformer |
| title_full | Automatic segmentation of white matter lesions on multi-parametric MRI: convolutional neural network versus vision transformer |
| title_fullStr | Automatic segmentation of white matter lesions on multi-parametric MRI: convolutional neural network versus vision transformer |
| title_full_unstemmed | Automatic segmentation of white matter lesions on multi-parametric MRI: convolutional neural network versus vision transformer |
| title_short | Automatic segmentation of white matter lesions on multi-parametric MRI: convolutional neural network versus vision transformer |
| title_sort | automatic segmentation of white matter lesions on multi parametric mri convolutional neural network versus vision transformer |
| topic | White matter hyperintensities Segmentation Brain MRI Convolutional neural network Vision transformer |
| url | https://doi.org/10.1186/s12883-024-04010-6 |
| work_keys_str_mv | AT yuntingchen automaticsegmentationofwhitematterlesionsonmultiparametricmriconvolutionalneuralnetworkversusvisiontransformer AT yanchenghuang automaticsegmentationofwhitematterlesionsonmultiparametricmriconvolutionalneuralnetworkversusvisiontransformer AT hsiulingchen automaticsegmentationofwhitematterlesionsonmultiparametricmriconvolutionalneuralnetworkversusvisiontransformer AT hsinchihlo automaticsegmentationofwhitematterlesionsonmultiparametricmriconvolutionalneuralnetworkversusvisiontransformer AT peichinchen automaticsegmentationofwhitematterlesionsonmultiparametricmriconvolutionalneuralnetworkversusvisiontransformer AT chiunchiehyu automaticsegmentationofwhitematterlesionsonmultiparametricmriconvolutionalneuralnetworkversusvisiontransformer AT yichintu automaticsegmentationofwhitematterlesionsonmultiparametricmriconvolutionalneuralnetworkversusvisiontransformer AT tyngluhliu automaticsegmentationofwhitematterlesionsonmultiparametricmriconvolutionalneuralnetworkversusvisiontransformer AT weichelin automaticsegmentationofwhitematterlesionsonmultiparametricmriconvolutionalneuralnetworkversusvisiontransformer |