MRSpineSeg Challenge

第二届中国图像与图形学会 (CSIG) 图像与图形技术挑战: MRSpineSeg挑战: 体积MR图像上脊柱结构的自动多类分割。
退行性脊柱疾病 (如腰椎间盘突出症、椎管狭窄症等) 已成为影响老年人和工作人群健康和生活质量的重要疾病。这些脊柱退行性病变常引起脊柱的结构形态和机械系统的改变，从而产生疼痛，如腰椎间盘突出症、椎间盘高度降低、神经压迫等。磁共振成像 (MRI) 作为一种无创性检查方法，具有良好的软组织成像和无辐射。它是退行性脊柱疾病的可靠筛查方法。在临床实践中，退行性脊柱疾病的治疗很大程度上取决于医师的经验，并且缺乏准确的定量分析工具。通过MRI对多类脊柱结构进行3D自动分割 (MR) 图像是对脊柱结构进行3D重建的前提。它可以为建立脊柱的生物力学模型，模拟脊柱结构中的应力以及评估退行性脊柱疾病的不同治疗方案的预后提供定量分析工具。
本次竞赛旨在聚集全球开发者，探索利用人工智能技术对MR图像中脊柱结构进行高效、准确的三维自动分割。脊柱结构应分段包括10个椎骨和9个椎间盘。

The 2nd China Society of Image and Graphics (CSIG) Image and Graphics Technology Challenge: MRSpineSeg Challenge: Automatic Multi-class Segmentation of Spinal Structures on Volumetric MR Images.

Degenerative spinal diseases, such as lumbar disc herniation and spinal stenosis, have become major threats to the health and quality of life of the elderly and working populations. These degenerative spinal disorders often induce alterations in the spinal structural morphology and mechanical system, leading to symptoms including pain, as well as conditions like lumbar disc herniation, reduced intervertebral disc height, and nerve compression. Magnetic Resonance Imaging (MRI), as a non-invasive examination method, boasts excellent soft tissue imaging capability and no radiation exposure, making it a reliable screening tool for degenerative spinal diseases. In clinical practice, the treatment of degenerative spinal diseases largely depends on physicians' experience, and there is a lack of accurate quantitative analysis tools. Automatic 3D segmentation of multiple spinal structures on MR images serves as a prerequisite for 3D reconstruction of spinal structures. It can provide quantitative analysis tools for establishing spinal biomechanical models, simulating stresses within spinal structures, and evaluating the prognosis of different treatment plans for degenerative spinal diseases.

This competition aims to gather global developers to explore the application of artificial intelligence technologies for efficient and accurate 3D automatic segmentation of spinal structures in MR images. The spinal structures to be segmented include 10 vertebrae and 9 intervertebral discs.

搜集汇总

背景与挑战

背景概述

MRSpineSeg Challenge是一个专注于医学图像分割的公开数据集，用于磁共振（MR）图像中脊柱结构的三维自动多类分割，包括10个椎骨和9个椎间盘。该数据集由南方医科大学于2021年发布，旨在通过人工智能技术辅助退行性脊柱疾病的定量分析和治疗评估，支持临床研究和生物力学模型构建。