Omni-Supervised Learning: Scaling Up to Large Unlabelled Medical Datasets

Two major bottlenecks in increasing algorithmic performance in the field of medical imaging analysis are the typically limited size of datasets and the shortage of expert labels for large datasets. This paper investigates approaches to overcome the latter via omni-supervised learning: a special case of semi-supervised learning. Our approach seeks to exploit a small annotated dataset and iteratively increase model performance by scaling up to refine the model using a large set of unlabelled data. By fusing predictions of perturbed inputs, the method generates new training annotations without human intervention. We demonstrate the effectiveness of the proposed framework to localize multiple structures in a 3D US dataset of 4044 fetal brain volumes with an initial expert annotation of just 200 volumes (5% in total) in training. Results show that structure localization error was reduced from 2.07 +/- 1.65mm to 1.76 +/- 1.35mm on the hold-out validation set.