Thoracic aortic dissection and aneurysms are the most lethal diseases of the aorta. The major hindrance to treatment lies in the accurate analysis of the medical images. More particularly, aortic segmentation of the 3Dimage is often tedious and difficult. Deep-learning-based segmentation models are an ideal solution, but their inability to deliver usable outputs in difficult cases and their computational cost cause their clinical adoptionto stay limited. This study presents an innovative approach for efficient aortic segmentation using targeted region of interest (ROI) detection. In contrast to classical detection models, we propose a simple and efficient detection model that can be widely applied to detect a single ROI. Our detection model is trained as a multi-task model, using an encoder-decoder architecture for segmentation and a fully connected network attached to the bottleneck for detection. We compare the performance of a one-step segmentation model applied to a complete image, nnU-Net and our cascade model composed of a detection and a segmentation step. We achieve a mean Dice similarity coefficient of 0.944 with over 0.9 for all cases using a third of the computing power. This simple solution achieves state-of-the-art performance while being compact and robust, making it an ideal solution for clinical applications.
Giordano, L, Dirks, I, Lenaerts, T & Vandemeulebroucke, J 2025, 'Region of interest detection for efficient aortic segmentation', Journal of medical imaging., vol. 13406, 13406-56. https://doi.org/10.1117/12.3046724
Giordano, L., Dirks, I., Lenaerts, T., & Vandemeulebroucke, J. (2025). Region of interest detection for efficient aortic segmentation. Journal of medical imaging., 13406, Article 13406-56. https://doi.org/10.1117/12.3046724
@article{1be8cc686b744227a548a8bd0b7caa99,
title = "Region of interest detection for efficient aortic segmentation",
abstract = "Thoracic aortic dissection and aneurysms are the most lethal diseases of the aorta. The major hindrance to treatment lies in the accurate analysis of the medical images. More particularly, aortic segmentation of the 3Dimage is often tedious and difficult. Deep-learning-based segmentation models are an ideal solution, but their inability to deliver usable outputs in difficult cases and their computational cost cause their clinical adoptionto stay limited. This study presents an innovative approach for efficient aortic segmentation using targeted region of interest (ROI) detection. In contrast to classical detection models, we propose a simple and efficient detection model that can be widely applied to detect a single ROI. Our detection model is trained as a multi-task model, using an encoder-decoder architecture for segmentation and a fully connected network attached to the bottleneck for detection. We compare the performance of a one-step segmentation model applied to a complete image, nnU-Net and our cascade model composed of a detection and a segmentation step. We achieve a mean Dice similarity coefficient of 0.944 with over 0.9 for all cases using a third of the computing power. This simple solution achieves state-of-the-art performance while being compact and robust, making it an ideal solution for clinical applications.",
keywords = "Detection, Segmentation, Multi-task learning, Cascade models, Aorta, Computed tomography",
author = "Loris Giordano and Ine Dirks and Tom Lenaerts and Jef Vandemeulebroucke",
note = "Publisher Copyright: {\textcopyright} 2025 SPIE.; SPIE 2025 - Medical Imaging : Image Processing ; Conference date: 16-02-2025 Through 20-02-2025",
year = "2025",
month = apr,
day = "11",
doi = "10.1117/12.3046724",
language = "English",
volume = "13406",
journal = "Journal of medical imaging.",
publisher = "SPIE",
url = "https://spie.org/conferences-and-exhibitions/medical-imaging",
}