Skip to main content
QUICK REVIEW

[论文解读] Towards contrast-agnostic soft segmentation of the spinal cord

Sandrine Bédard, Enamundram Naga Karthik|arXiv (Cornell University)|Oct 23, 2023
Medical Imaging and Analysis被引用 6
一句话总结

本论文提出一种无需依赖对比度的深度学习方法,用于使用多对比度MRI数据进行软脊髓分割,通过在六个对比度上平均获得个体特异的软真实标签,并使用回归损失训练UNet模型。该方法降低了跨对比度的截面积(CSA)变异性,并在不同厂商、对比度和病理情况(如病灶、脊髓压迫)下表现出更优的泛化能力,同时考虑了部分体积效应。

ABSTRACT

Spinal cord segmentation is clinically relevant and is notably used to compute spinal cord cross-sectional area (CSA) for the diagnosis and monitoring of cord compression or neurodegenerative diseases such as multiple sclerosis. While several semi and automatic methods exist, one key limitation remains: the segmentation depends on the MRI contrast, resulting in different CSA across contrasts. This is partly due to the varying appearance of the boundary between the spinal cord and the cerebrospinal fluid that depends on the sequence and acquisition parameters. This contrast-sensitive CSA adds variability in multi-center studies where protocols can vary, reducing the sensitivity to detect subtle atrophies. Moreover, existing methods enhance the CSA variability by training one model per contrast, while also producing binary masks that do not account for partial volume effects. In this work, we present a deep learning-based method that produces soft segmentations of the spinal cord. Using the Spine Generic Public Database of healthy participants ($ ext{n}=267$; $ ext{contrasts}=6$), we first generated participant-wise soft ground truth (GT) by averaging the binary segmentations across all 6 contrasts. These soft GT, along with aggressive data augmentation and a regression-based loss function, were used to train a U-Net model for spinal cord segmentation. We evaluated our model against state-of-the-art methods and performed ablation studies involving different loss functions and domain generalization methods. Our results show that using the soft segmentations along with a regression loss function reduces CSA variability ($p < 0.05$, Wilcoxon signed-rank test). The proposed spinal cord segmentation model generalizes better than the state-of-the-art methods amongst unseen datasets, vendors, contrasts, and pathologies (compression, lesions), while accounting for partial volume effects.

研究动机与目标

  • 解决现有分割方法中因MRI对比度依赖性导致的脊髓截面积(CSA)测量变异性问题。
  • 通过在多对比度生成的软真实标签上训练单一模型,减少不同对比度间的CSA变异性。
  • 通过利用多样化训练数据,提升模型在未见MRI厂商、对比度和病理情况(如病灶、压迫)下的泛化能力。
  • 通过软分割方法考虑部分体积效应,避免二值掩码忽略组织边界不确定性的问题。
  • 为临床和多中心研究中稳健的多对比度脊髓形态测量建立基础模型。

提出的方法

  • 从Spine Generic Public Database(n=267)中,基于六种MRI对比度(T1w、T2w、T2w-SPACE、T2w-TSE、T2w-TSE-SPACE、T2w-TSE-SPACE-3D)的二值分割结果,通过平均生成个体特异的软真实标签。
  • 应用激进的数据增强策略(旋转、翻转、缩放、强度抖动),以提升模型的鲁棒性和泛化能力。
  • 采用基于回归的损失函数训练基于UNet的深度学习模型,以预测软分割掩码而非二值掩码。
  • 在预处理中采用中心裁剪策略,假设脊髓位于图像中心位置,并通过可配置标志处理特殊情况(如脊柱侧弯、腰椎区域)。
  • 使用Wilcoxon符号秩检验评估CSA变异性,并与最先进对比度特异性模型(如soft_per_contrast T2w、DeepSeg2D)进行比较。
  • 通过消融实验,改变真实标签类型、损失函数和对比度特异性模型,以分离软真实标签和回归损失的贡献。
Figure 1: Preprocessing pipeline for soft average segmentations ground truth. (1) Automatic hard spinal cord segmentation using sct_deepseg_sc & manual corrections; (2) Registration to T2w space; (3) Applying each contrast’s warping field to bring the segmentation masks to the T2w space; (4) Averagi
Figure 1: Preprocessing pipeline for soft average segmentations ground truth. (1) Automatic hard spinal cord segmentation using sct_deepseg_sc & manual corrections; (2) Registration to T2w space; (3) Applying each contrast’s warping field to bring the segmentation masks to the T2w space; (4) Averagi

实验结果

研究问题

  • RQ1能否通过多对比度生成的软真实标签,降低不同MRI序列间的CSA变异性?
  • RQ2与二值交叉熵损失相比,基于回归的损失函数是否能提升模型在未见MRI对比度和厂商下的泛化能力?
  • RQ3在多对比度数据上训练的单一深度学习模型,能否泛化到未见的病理情况(如病灶和脊髓压迫)?
  • RQ4与二值分割相比,软分割在多大程度上缓解了部分体积效应?
  • RQ5在CSA一致性与泛化性能方面,该方法与对比度特异性模型相比表现如何?

主要发现

  • 采用软真实标签和回归损失的模型,在所有六个对比度上显著降低了CSA变异性(p < 0.05,Wilcoxon符号秩检验)。
  • 该模型在未见数据集、厂商、对比度和病理情况(包括病灶和脊髓压迫)下,泛化能力优于最先进对比度特异性方法。
  • 软分割提升了边界勾勒能力,尤其在存在病灶时优于单对比度训练的模型。
  • 消融实验证实,软真实标签和回归损失是降低CSA变异性与提升泛化能力的关键因素。
  • 该模型在未见对比度和病理情况中表现出稳健性能,表明其具备强大的零样本泛化潜力。
  • 该方法有望通过最小化对比度引起的形态测量偏差,提升对细微脊髓萎缩的检测敏感性。
Figure 2: Architecture of the proposed SoftSeg model.
Figure 2: Architecture of the proposed SoftSeg model.

更好的研究,从现在开始

从阅读论文到最终审阅,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。