Benchmarking Geometric Deep Learning for Cortical Segmentation and Neurodevelopmental Phenotype Prediction
Benchmarking Geometric Deep Learning for Cortical Segmentation and Neurodevelopmental Phenotype Prediction
复制标题
DOI：
10.1101/2021.12.01.470730
复制
发表时间：
2021-12
期刊：
影响因子：
--
通讯作者：
Abdulah Fawaz;Logan Z. J. Williams;A. Alansary;Cher Bass;Karthik Gopinath;Mariana da Silva;Simon Dahan-Simo
中科院分区：
文献类型：
--
作者：
Abdulah Fawaz;Logan Z. J. Williams;A. Alansary;Cher Bass;Karthik Gopinath;Mariana da Silva;Simon Dahan-Simo
新兴的几何深度学习领域将卷积神经网络的应用扩展到不规则的领域，如图形、网格和曲面。最近的几项研究探索了使用这些技术分析和分割皮质表面的潜力。然而，到目前为止，还没有对这些方法进行全面的比较，也没有对现有的欧几里得方法进行比较。本文对来自公开可用的人类连接体发展项目（dHCP）的球形新生儿皮质表面数据的表型预测和分割的几何和传统深度学习模型集合进行了基准测试。任务包括预测扫描时的月经后年龄，出生时的胎龄，以及通过M-CRIB-S图谱将皮质表面划分为解剖区域。不仅根据模型精度评估性能，还根据网络对图像配准的依赖以及通过遮挡进行模型解释来评估性能。网络在球面化和解剖皮质网格上进行训练。研究结果表明，几何深度学习相对于传统深度学习的效用是高度任务特异性的，这对未来皮层表面深度学习模型的设计具有重要意义。数据访问的代码和指令可从https://github.com/Abdulah-Fawaz/Benchmarking-Surface-DL获得。
The emerging field of geometric deep learning extends the application of convolutional neural networks to irregular domains such as graphs, meshes and surfaces. Several recent studies have explored the potential for using these techniques to analyse and segment the cortical surface. However, there has been no comprehensive comparison of these approaches to one another, nor to existing Euclidean methods, to date. This paper benchmarks a collection of geometric and traditional deep learning models on phenotype prediction and segmentation of sphericalised neonatal cortical surface data, from the publicly available Developing Human Connectome Project (dHCP). Tasks include prediction of postmenstrual age at scan, gestational age at birth and segmentation of the cortical surface into anatomical regions defined by the M-CRIB-S atlas. Performance was assessed not only in terms of model precision, but also in terms of network dependence on image registration, and model interpretation via occlusion. Networks were trained both on sphericalised and anatomical cortical meshes. Findings suggest that the utility of geometric deep learning over traditional deep learning is highly task-specific, which has implications for the design of future deep learning models on the cortical surface. The code, and instructions for data access, are available from https://github.com/Abdulah-Fawaz/Benchmarking-Surface-DL.