Robust and generalizable segmentation of human functional tissue units

Abstract

The Human BioMolecular Atlas Program aims to compile a reference atlas for the healthy human adult body at the cellular level. Functional tissue units (FTU, e.g., renal glomeruli and colonic crypts) are of pathobiological significance and relevant for modeling and understanding disease progression. Yet, annotation of FTUs is time consuming and expensive when done manually and existing algorithms achieve low accuracy and do not generalize well. This paper compares the five winning algorithms from the “Hacking the Kidney” Kaggle competition to which more than a thousand teams from sixty countries contributed. We compare the accuracy and performance of the algorithms on a large-scale renal glomerulus Periodic acid-Schiff stain dataset and their generalizability to a colonic crypts hematoxylin and eosin stain dataset. Results help to characterize how the number of FTUs per unit area differs in relationship to their position in kidney and colon with respect to age, sex, body mass index (BMI), and other clinical data and are relevant for advancing pathology, anatomy, and surgery.

Spatially resolved proteomics via tissue expansion

“…The improved spatial resolution extends applicability of ProteomEx as it is more biologically relevant for analysis of heterogeneity of biological tissues. For example, ~100 µm resolution is more appropriate for analysis of functional tissue units (FTU) 38 , e.g., renal glomeruli [39][40][41] , colonic crypts 39 , human lung blood vessels 36 , which are typically limited by 100 µm in size. Furthermore, Piehowski et al demonstrated that 100 µm resolution was sufficient to characterize region-specific bioactivity and unique tissue microenvironments within the mouse Wnt5a-null uterine tissue 30 .…”

Segmenting functional tissue units across human organs using community-driven development of generalizable machine learning algorithms

“…In 2021, the Human BioMolecular Atlas Program (HuBMAP) conducted a Kaggle competition 10,11 that focussed exclusively on segmentation of renal glomeruli in PAS stained histological images of kidney tissue, engaging 1,200 teams from 60 countries.…”

“…In 2021, the Human BioMolecular Atlas Program (HuBMAP) conducted a Kaggle competition 10,11 that focussed exclusively on segmentation of renal glomeruli in PAS stained histological images of kidney tissue, engaging 1,200 teams from 60 countries. The winning model from this competition was validated and productized in the HuBMAP data portal; it is now being run on all kidney tissue data at scale.…”