基于频率分解学习的定量磁化率图中脑结构分割
Frequency Disentangled Learning for Segmentation of Midbrain Structures from Quantitative Susceptibility Mapping Data
摘要:
人们通常缺乏足够的注释样本来训练深度分割模型。对于不太常见的成像模态,例如定量敏感性标测(QSM),情况尤其如此。已经表明,深度模型倾向于拟合从低频到高频的目标函数。人们可以假设,可以利用这种属性来更好地训练深度学习模型。在本文中,我们利用这个属性,提出了一种新的训练方法的基础上频域解纠缠。它包括两个主要步骤:i)将图像分解为高频和低频部分并进行特征学习; ii)频域融合以完成任务。该方法可用于任何骨干分段网络。我们应用的方法,从QSM数据,这是特别相关的帕金森综合征的研究的红色和齿状核的分割。我们证明,所提出的方法为这些任务提供了相当大的性能改进。我们进一步将其应用于医学细分十项全能(MSD)挑战的三个公共数据集。对于两个MSD任务,它提供了较小但仍然很大的改进(高达7点Dice),特别是在小训练集的情况下。
英文摘要:
One often lacks sufficient annotated samples for training deep segmentation models. This is in particular the case for less common imaging modalities such as Quantitative Susceptibility Mapping (QSM). It has been shown that deep models tend to fit the target function from low to high frequencies. One may hypothesize that such property can be leveraged for better training of deep learning models. In this paper, we exploit this property to propose a new training method based on frequency-domain disentanglement. It consists of two main steps: i) disentangling the image into high- and low-frequency parts and feature learning; ii) frequency-domain fusion to complete the task. The approach can be used with any backbone segmentation network. We apply the approach to the segmentation of the red and dentate nuclei from QSM data which is particularly relevant for the study of parkinsonian syndromes. We demonstrate that the proposed method provides considerable performance improvements for these tasks. We further applied it to three public datasets from the Medical Segmentation Decathlon (MSD) challenge. For two MSD tasks, it provided smaller but still substantial improvements (up to 7 points of Dice), especially under small training set situations.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。