1
|
Lin H, López-Tapia S, Schiffers F, Wu Y, Gunasekaran S, Hwang J, Bishara D, Kholmovski E, Elbaz M, Passman RS, Kim D, Katsaggelos AK. Usformer: A small network for left atrium segmentation of 3D LGE MRI. Heliyon 2024; 10:e28539. [PMID: 38596055 PMCID: PMC11002571 DOI: 10.1016/j.heliyon.2024.e28539] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/14/2024] [Accepted: 03/20/2024] [Indexed: 04/11/2024] Open
Abstract
Left atrial (LA) fibrosis plays a vital role as a mediator in the progression of atrial fibrillation. 3D late gadolinium-enhancement (LGE) MRI has been proven effective in identifying LA fibrosis. Image analysis of 3D LA LGE involves manual segmentation of the LA wall, which is both lengthy and challenging. Automated segmentation poses challenges owing to the diverse intensities in data from various vendors, the limited contrast between LA and surrounding tissues, and the intricate anatomical structures of the LA. Current approaches relying on 3D networks are computationally intensive since 3D LGE MRIs and the networks are large. Regarding this issue, most researchers came up with two-stage methods: initially identifying the LA center using a scaled-down version of the MRIs and subsequently cropping the full-resolution MRIs around the LA center for final segmentation. We propose a lightweight transformer-based 3D architecture, Usformer, designed to precisely segment LA volume in a single stage, eliminating error propagation associated with suboptimal two-stage training. The transposed attention facilitates capturing the global context in large 3D volumes without significant computation requirements. Usformer outperforms the state-of-the-art supervised learning methods in terms of accuracy and speed. First, with the smallest Hausdorff Distance (HD) and Average Symmetric Surface Distance (ASSD), it achieved a dice score of 93.1% and 92.0% in the 2018 Atrial Segmentation Challenge and our local institutional dataset, respectively. Second, the number of parameters and computation complexity are largely reduced by 2.8x and 3.8x, respectively. Moreover, Usformer does not require a large dataset. When only 16 labeled MRI scans are used for training, Usformer achieves a 92.1% dice score in the challenge dataset. The proposed Usformer delineates the boundaries of the LA wall relatively accurately, which may assist in the clinical translation of LA LGE for planning catheter ablation of atrial fibrillation.
Collapse
Affiliation(s)
- Hui Lin
- Department of Electrical and Computer Engineering, Northwestern University, Evanston, IL, USA
| | - Santiago López-Tapia
- Department of Electrical and Computer Engineering, Northwestern University, Evanston, IL, USA
| | - Florian Schiffers
- Department of Electrical and Computer Engineering, Northwestern University, Evanston, IL, USA
| | - Yunan Wu
- Department of Electrical and Computer Engineering, Northwestern University, Evanston, IL, USA
| | | | - Julia Hwang
- Department of Radiology, Northwestern University, Chicago, IL, USA
| | - Dima Bishara
- Department of Radiology, Northwestern University, Chicago, IL, USA
| | - Eugene Kholmovski
- Department of Biomedical Engineering, Johns Hopkins University, Maryland, USA
| | - Mohammed Elbaz
- Department of Radiology, Northwestern University, Chicago, IL, USA
- Department of Biomedical Engineering, Northwestern University, Evanston, IL, USA
| | - Rod S. Passman
- Department of Medicine, Northwestern University, Chicago, IL, USA
| | - Daniel Kim
- Department of Radiology, Northwestern University, Chicago, IL, USA
- Department of Biomedical Engineering, Northwestern University, Evanston, IL, USA
| | - Aggelos K. Katsaggelos
- Department of Electrical and Computer Engineering, Northwestern University, Evanston, IL, USA
| |
Collapse
|