Liu Xiaofeng, Xing Fangxu, Marin Thibault, Fakhri Georges El, Woo Jonghye
Gordon Center for Medical Imaging, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114 USA.
Proc SPIE Int Soc Opt Eng. 2022 Feb-Mar;12032. doi: 10.1117/12.2604547. Epub 2022 Apr 4.
Lesions or organ boundaries visible through medical imaging data are often ambiguous, thus resulting in significant variations in multi-reader delineations, i.e., the source of aleatoric uncertainty. In particular, quantifying the inter-observer variability of manual annotations with Magnetic Resonance (MR) Imaging data plays a crucial role in establishing a reference standard for various diagnosis and treatment tasks. Most segmentation methods, however, simply model a mapping from an image to its single segmentation map and do not take the disagreement of annotators into consideration. In order to account for inter-observer variability, without sacrificing accuracy, we propose a novel variational inference framework to model the distribution of plausible segmentation maps, given a specific MR image, which explicitly represents the multi-reader variability. Specifically, we resort to a latent vector to encode the multi-reader variability and counteract the inherent information loss in the imaging data. Then, we apply a variational autoencoder network and optimize its evidence lower bound (ELBO) to efficiently approximate the distribution of the segmentation map, given an MR image. Experimental results, carried out with the QUBIQ brain growth MRI segmentation datasets with seven annotators, demonstrate the effectiveness of our approach.
通过医学成像数据可见的病变或器官边界往往不明确，从而导致多位阅片者的描绘存在显著差异，即偶然不确定性的来源。特别是，量化磁共振（MR）成像数据手动标注的观察者间变异性在为各种诊断和治疗任务建立参考标准方面起着关键作用。然而，大多数分割方法只是简单地对从图像到其单一分割图的映射进行建模，而没有考虑标注者之间的分歧。为了在不牺牲准确性的情况下考虑观察者间变异性，我们提出了一种新颖的变分推理框架，以对给定特定MR图像的合理分割图分布进行建模，该框架明确表示了多位阅片者的变异性。具体来说，我们借助一个潜在向量来编码多位阅片者的变异性，并抵消成像数据中固有的信息损失。然后，我们应用一个变分自编码器网络并优化其证据下界（ELBO），以在给定MR图像的情况下有效地近似分割图的分布。使用有七位标注者的QUBIQ脑生长MRI分割数据集进行的实验结果证明了我们方法的有效性。
