P-Count: Persistence-based Counting of White Matter Hyperintensities in Brain MRI
This addresses a specific challenge in neuroimaging for researchers and clinicians by providing a more reliable lesion count, though it is an incremental improvement over existing segmentation methods.
The authors tackled the problem of accurately counting white matter hyperintensities (WMH) in brain MRI, which is sensitive to noise and segmentation errors, by developing P-Count, a persistence-based method that filters out noisy positives, resulting in significantly more accurate counts on the ISBI2015 dataset.
White matter hyperintensities (WMH) are a hallmark of cerebrovascular disease and multiple sclerosis. Automated WMH segmentation methods enable quantitative analysis via estimation of total lesion load, spatial distribution of lesions, and number of lesions (i.e., number of connected components after thresholding), all of which are correlated with patient outcomes. While the two former measures can generally be estimated robustly, the number of lesions is highly sensitive to noise and segmentation mistakes -- even when small connected components are eroded or disregarded. In this article, we present P-Count, an algebraic WMH counting tool based on persistent homology that accounts for the topological features of WM lesions in a robust manner. Using computational geometry, P-Count takes the persistence of connected components into consideration, effectively filtering out the noisy WMH positives, resulting in a more accurate count of true lesions. We validated P-Count on the ISBI2015 longitudinal lesion segmentation dataset, where it produces significantly more accurate results than direct thresholding.