Comparative evaluation of training strategies using partially labelled datasets for segmentation of white matter hyperintensities and stroke lesions in FLAIR MRI
This work addresses the difficulty in developing deep learning models for medical image segmentation in cerebral small vessel disease, but it is incremental as it focuses on comparing existing training strategies rather than introducing a new paradigm.
The study tackled the challenge of segmenting white matter hyperintensities and stroke lesions in FLAIR MRI, which are visually confounding, by evaluating six training strategies using partially labelled datasets on 2052 MRI volumes, finding that pseudolabels yielded the best performance improvement.
White matter hyperintensities (WMH) and ischaemic stroke lesions (ISL) are imaging features associated with cerebral small vessel disease (SVD) that are visible on brain magnetic resonance imaging (MRI) scans. The development and validation of deep learning models to segment and differentiate these features is difficult because they visually confound each other in the fluid-attenuated inversion recovery (FLAIR) sequence and often appear in the same subject. We investigated six strategies for training a combined WMH and ISL segmentation model using partially labelled data. We combined privately held fully and partially labelled datasets with publicly available partially labelled datasets to yield a total of 2052 MRI volumes, with 1341 and 1152 containing ground truth annotations for WMH and ISL respectively. We found that several methods were able to effectively leverage the partially labelled data to improve model performance, with the use of pseudolabels yielding the best result.