Curriculum semi-supervised segmentation
This addresses efficient use of unlabeled data for medical image segmentation, but it is incremental as it builds on existing semi-supervision methods.
The study tackled semi-supervised segmentation by using a curriculum-style strategy with a regression network to learn image-level information like target region size, which regularizes the segmentation network to match inferred label distributions, achieving results approaching full-supervision for left ventricle MRI segmentation.
This study investigates a curriculum-style strategy for semi-supervised CNN segmentation, which devises a regression network to learn image-level information such as the size of a target region. These regressions are used to effectively regularize the segmentation network, constraining softmax predictions of the unlabeled images to match the inferred label distributions. Our framework is based on inequality constraints that tolerate uncertainties with inferred knowledge, e.g., regressed region size, and can be employed for a large variety of region attributes. We evaluated our proposed strategy for left ventricle segmentation in magnetic resonance images (MRI), and compared it to standard proposal-based semi-supervision strategies. Our strategy leverages unlabeled data in more efficiently, and achieves very competitive results, approaching the performance of full-supervision.