ZOZI-Seg: A transformer and UNet cascade network with Zoom-Out and Zoom-In scheme for aortic dissection segmentation in enhanced CT images

• Published in: Computers in biology and medicine Vol. 175; p. 108494
• Main Authors: Jung, Ji-Hoon, Oh, Hong Min, Jeong, Gyu-Jun, Kim, Tae-Won, Koo, Hyun Jung, Lee, June-Goo, Yang, Dong Hyun
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.06.2024; Elsevier Limited
• Subjects: Ablation; Accuracy; Algorithms; Aorta; Aorta segmentation; Aortic dissection; Aortic Dissection - diagnostic imaging; Computed tomography; Context; Coronary vessels; Datasets; Deep learning; Diagnosis; Dissection; Health care facilities; Humans; Image enhancement; Image processing; Image segmentation; Medical diagnosis; Medical imaging; nnUNet; Segmentation; Texture; Thromboembolism; Thrombosis; Tomography, X-Ray Computed - methods; Transformer; Transformers; U-Net; CT; Aortic dissection; Transformer; nnUNet; Aorta segmentation; U-Net
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/j.compbiomed.2024.108494

Aortic dissection (AD) is a serious condition requiring rapid and accurate diagnosis. In this study, we aimed to improve the diagnostic accuracy of AD by presenting a novel method for aortic segmentation in computed tomography images that uses a combination of a transformer and a UNet cascade network with a Zoom-Out and Zoom-In scheme (ZOZI-seg). The proposed method segments each compartment of the aorta, comprising the true lumen (TL), false lumen (FL), and thrombosis (TH) using a cascade strategy that captures both the global context (anatomical structure) and the local detail texture based on the dynamic patch size with ZOZI schemes. The ZOZI-seg model has a two-stage architecture using both a “3D transformer for panoptic context-awareness” and a “3D UNet for localized texture refinement.” The unique ZOZI strategies for patching were demonstrated in an ablation study. The performance of our proposed ZOZI-seg model was tested using a dataset from Asan Medical Center and compared with those of existing models such as nnUNet and nnFormer. In terms of segmentation accuracy, our method yielded better results, with Dice similarity coefficients (DSCs) of 0.917, 0.882, and 0.630 for TL, FL, and TH, respectively. Furthermore, we indirectly compared our model with those in previous studies using an external dataset to evaluate its robustness and generalizability. This approach may help in the diagnosis and treatment of AD in different clinical situations and provide a strong basis for further research and clinical applications. •We address precise Aortic Dissection (AD) segmentation in CT with a novel model ZOZI-seg for effective diagnosis and treatment.•ZOZI-seg uses two-stage cascade method combining CNN and Transformer to capture global context and local texture details.•Unique Zoom-Out Zoom-In scheme applies different patch sizes: 3D Transformer for overall structure, 3D UNet for texture refinement.•ZOZI-seg's superior performance suggests potential to enhance diagnostic accuracy and treatment strategies for AD patients.