Adapting Off-the-Shelf Source Segmenter for Target Medical Image Segmentation

Abstract

Unsupervised domain adaptation (UDA) aims to transfer knowledge learned from a labeled source domain to an unlabeled and unseen target domain, which is usually trained on data from both domains. Access to the source domain data at the adaptation stage, however, is often limited, due to data storage or privacy issues. To alleviate this, in this work, we target source free UDA for segmentation, and propose to adapt an "off-the-shelf" segmentation model pre-trained in the source domain to the target domain, with an adaptive batch-wise normalization statistics adaptation framework. Specifically, the domain-specific low-order batch statistics, i.e., mean and variance, are gradually adapted with an exponential momentum decay scheme, while the consistency of domain shareable high-order batch statistics, i.e., scaling and shifting parameters, is explicitly enforced by our optimization objective. The transferability of each channel is adaptively measured first from which to balance the contribution of each channel. Moreover, the proposed source free UDA framework is orthogonal to unsupervised learning methods, e.g., self-entropy minimization, which can thus be simply added on top of our framework. Extensive experiments on the BraTS 2018 database show that our source free UDA framework outperformed existing source-relaxed UDA methods for the cross-subtype UDA segmentation task and yielded comparable results for the cross-modality UDA segmentation task, compared with a supervised UDA methods with the source data.

Variational Inference for Quantifying Inter-observer Variability in Segmentation of Anatomical Structures

“…[10][11][12][13][14][15] The semantic segmentation makes pixelwise classification and provides more explainable results for the subsequent decision making. 16,17 However, the conventional training of segmentation neural networks utilized a single label as the ground-truth, and did not consider the inter-observer variability.…”

Generative Self-training for Cross-domain Unsupervised Tagged-to-Cine MRI Synthesis

“…In addition, due to the ability of deep learning to disentangle explanatory factors of variations, efforts have been made to learn more transferable features. Recent works in UDA incorporated discrepancy measures into network architectures to align feature distributions between source and target domains [18,19]. This was achieved by either minimizing the distribution discrepancy between feature distribution statistics, e.g., maximum mean discrepancy (MMD), or adversarially learning the feature representations to fool a domain classifier in a two-player minimax game [18].…”

Adversarial Unsupervised Domain Adaptation with Conditional and Label Shift: Infer, Align and Iterate

“…The real world implementation scenarios, however, can be significantly diverse, and it can be costly to label datasets in every target environment [10,41]. To address this, unsupervised domain adaptation (UDA) can be used to transfer knowledge learned from a labeled source domain to different unlabeled target domains [13,33,32].…”