Multi-Class Segmentation of Aortic Branches and Zones in Computed Tomography Angiography: The AortaSeg24 Challenge

2025-02-07

AI Summary

Existing aortic segmentation methods are limited to binary classification, hindering clinical measurement of branch diameters and regional anatomical analysis, and suffer from a lack of open-source multi-class datasets. To address this, we introduce the first open-source multi-class aortic CTA dataset—comprising 100 cases annotated with 23 anatomical structures—and establish the first standardized multi-class segmentation benchmark, enabling a paradigm shift from binary to anatomy-aware fine-grained segmentation. We propose a clinically motivated joint evaluation metric (Dice + Normalized Surface Distance) and develop a nnU-Net–based cascade architecture incorporating a customized loss function, multi-scale augmentation, and surface-aware optimization. Our framework attracted 121 teams in an international challenge, with top-performing models achieving an average Dice score exceeding 0.82. All data, source code, and winning models are publicly released.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Multi-class segmentation of the aorta in computed tomography angiography (CTA) scans is essential for diagnosing and planning complex endovascular treatments for patients with aortic dissections. However, existing methods reduce aortic segmentation to a binary problem, limiting their ability to measure diameters across different branches and zones. Furthermore, no open-source dataset is currently available to support the development of multi-class aortic segmentation methods. To address this gap, we organized the AortaSeg24 MICCAI Challenge, introducing the first dataset of 100 CTA volumes annotated for 23 clinically relevant aortic branches and zones. This dataset was designed to facilitate both model development and validation. The challenge attracted 121 teams worldwide, with participants leveraging state-of-the-art frameworks such as nnU-Net and exploring novel techniques, including cascaded models, data augmentation strategies, and custom loss functions. We evaluated the submitted algorithms using the Dice Similarity Coefficient (DSC) and Normalized Surface Distance (NSD), highlighting the approaches adopted by the top five performing teams. This paper presents the challenge design, dataset details, evaluation metrics, and an in-depth analysis of the top-performing algorithms. The annotated dataset, evaluation code, and implementations of the leading methods are publicly available to support further research. All resources can be accessed at https://aortaseg24.grand-challenge.org.

Problem

• Multi-class aortic segmentation in CTA scans
• Lack of open-source dataset for development
• Evaluation of top-performing segmentation algorithms

Innovation

• Multi-class aortic segmentation dataset
• Utilized nnU-Net and cascaded models
• Publicly accessible evaluation metrics