Spatio-temporal Consistency and Negative Label Transfer for 3D freehand US Segmentation

The manual segmentation of multiple organs in 3D ultra-sound (US) sequences and volumes towards their quantitative analysis is very expensive and time-consuming. Fully supervised segmenta-tion methods still require the collection of large volumes of annotated data while unlabeled images are abundant. In this work, we propose a semi-automatic deep learning approach modeled as a weak-label learning problem: given a few 2-D incomplete annotations for selected slices, the goal is to propagate the masks to the entire sequence. To this end, we make use of both positive and negative constraints induced by incomplete labels to penalize the segmentation loss function. Our model is composed of one encoder and two decoders to model the segmentation and an auxiliary reconstruction task. Moreover, we consider the spatio-temporal information by deploying a Convolutional Long Short Term Memory module. Our findings suggest that the reconstruction decoder and the spatio-temporal information lead to a better geometrical estimation of the mask shape. We apply the model to the task of low-limb muscle segmentation in a dataset of 44 patients and 6160 images.