用物理约束生成MRI数据，提升脑卒中病灶分割的跨域泛化能力。
Domain-Agnostic Stroke Lesion Segmentation Using Physics-Constrained Synthetic Data
- 基于MRI物理规律生成合成qMRI图像，确保组织对比真实
- 在多个跨域数据集上分割精度超越基线UNet模型
- 适合需要高鲁棒性医学图像分割的研究者使用
由于不同扫描协议导致的数据异质性，MRI中脑卒中病灶分割面临挑战，限制了模型的泛化能力。本文提出两种物理约束的合成方法： 1. $ exttt{qATLAS}$：训练神经网络从标准MPRAGE图像估计定量MRI（qMRI）图，实现多种序列的逼真模拟； 2. $ exttt{qSynth}$：基于组织标签，使用条件高斯混合模型直接合成物理合理的qMRI图。 在多个跨域数据集上的实验表明，两种方法均优于基线UNet，其中$ exttt{qSynth}$显著超越以往合成数据方法。结果验证了将MRI物理知识融入合成数据生成对提升分割鲁棒性与泛化性的潜力。代码已公开于https://github.com/liamchalcroft/qsynth。
原文摘要 · Abstract (English)
Segmenting stroke lesions in MRI is challenging due to diverse acquisition protocols that limit model generalisability. In this work, we introduce two physics-constrained approaches to generate synthetic quantitative MRI (qMRI) images that improve segmentation robustness across heterogeneous domains. Our first method, $\texttt{qATLAS}$, trains a neural network to estimate qMRI maps from standard MPRAGE images, enabling the simulation of varied MRI sequences with realistic tissue contrasts. The second method, $\texttt{qSynth}$, synthesises qMRI maps directly from tissue labels using label-conditioned Gaussian mixture models, ensuring physical plausibility. Extensive experiments on multiple out-of-domain datasets show that both methods outperform a baseline UNet, with $\texttt{qSynth}$ notably surpassing previous synthetic data approaches. These results highlight the promise of integrating MRI physics into synthetic data generation for robust, generalisable stroke lesion segmentation. Code is available at https://github.com/liamchalcroft/qsynth
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。