Pseudo-label refinement using superpixels for semi-supervised brain tumour segmentation
This addresses the challenge of reducing annotator burden for radiologists in medical imaging by enhancing semi-supervised segmentation, though it is incremental as it refines an existing technique.
The paper tackles the problem of inaccurate pseudo-labels degrading performance in semi-supervised brain tumour segmentation with limited annotations, proposing a superpixel-based refinement method that improves over the baseline, achieving DSC scores of 0.824 and 0.707 for whole tumour and tumour core regions on a test set with only 5 annotated patients.
Training neural networks using limited annotations is an important problem in the medical domain. Deep Neural Networks (DNNs) typically require large, annotated datasets to achieve acceptable performance which, in the medical domain, are especially difficult to obtain as they require significant time from expert radiologists. Semi-supervised learning aims to overcome this problem by learning segmentations with very little annotated data, whilst exploiting large amounts of unlabelled data. However, the best-known technique, which utilises inferred pseudo-labels, is vulnerable to inaccurate pseudo-labels degrading the performance. We propose a framework based on superpixels - meaningful clusters of adjacent pixels - to improve the accuracy of the pseudo labels and address this issue. Our framework combines superpixels with semi-supervised learning, refining the pseudo-labels during training using the features and edges of the superpixel maps. This method is evaluated on a multimodal magnetic resonance imaging (MRI) dataset for the task of brain tumour region segmentation. Our method demonstrates improved performance over the standard semi-supervised pseudo-labelling baseline when there is a reduced annotator burden and only 5 annotated patients are available. We report DSC=0.824 and DSC=0.707 for the test set whole tumour and tumour core regions respectively.