Liver Shape Analysis using Statistical Parametric Maps at Population Scale

Abstract

Morphometric image analysis enables the quantification of differences in the shape and size of organs between individuals. Here we have applied morphometric methods to the study of the liver by constructing surface meshes from liver segmentations from abdominal MRI images in 33,434 participants in the UK Biobank. Based on these three dimensional mesh vertices, we evaluated local shape variations and modelled their association with anthropometric, phenotypic and disease traits, including liver disease and type-2 diabetes. We found that age, body mass index, hepatic fat and iron content, as well as, health traits were significantly associated with regional liver shape and size. Interaction models in disease groups showed that the presence of type-2 diabetes accelerates age-related changes in the liver, while presence of liver fat further increased shape variation in both type-2 diabetes and liver disease. The results suggest that this novel approach may greatly benefit studies aiming at better categorisation of pathologies associated with acute and chronic liver conditions.

Kidney Shape Statistical Analysis: Associations with Disease and Anthropometric Factors

“…The registration process has been previously detailed in [23]. In brief, we used rigid registration to remove the position and orientation difference between all participant-specific surfaces and template surfaces.…”

“…To determine the regional outward or inward adaptations in kidney surface in comparison to an average kidney shape for each participant, the surface-to-surface (S2S) distance was measured by computing the signed distance between the template surface and the participants' surface mesh at each vertex [23].…”