Unsupervised Computational Labeling of Glomerular Textural Boundaries in Renal Pathology

Diagnosing renal disease requires time consuming manual inspection of needle biopsies. Automated quantification of the same characteristics of biopsies by a digital method greatly decreases the burden on pathologists and improves the reproducibility of the biopsy process. The current barrier to the automated quantification of renal injury in proteinuria is the digital identification of the glomerulus. We have developed an integrated method, based on Gabor filter bank based textural segmentation, statistical F-testing, and distance transform, for segmenting glomerular boundaries from renal biopsies. Our method outperforms sole Gabor filter bank based method, and is able to operate on multiple histological stains.

Computational pipeline for segmenting the glomerular boundary.
Computational pipeline for segmenting the glomerular boundary. (A) Hematoxylin and eosin stained glomerular tissue image. (B) Grayscale version of the image shown in A. (C) Gaussian blurred image of B. (D) Intensity image of the first principal component of the Gabor filter bank outputs using as input the image shown in C. (E) K-means clustering was used to find final Gabor boundary. (F) F-testing examines the entire image for similarity with E, and outputs 0 or 1 for each pixel. (G) Morphological noise removal for the image in F. (H–I) Respective binary masks obtained from F-testing and Gabor filter bank were distance transformed. (J) Intensity image of a spatial weighting intensity map obtained from A. (K) Heatmap of an average of intensity images in H–J. (L) Final segmentation after thresholding shown using green. The segmentation obtained from initial Gabor (see E) is shown using black. The green boundary depicts improved detection of glomerulus.

Select References

B. Ginley, J. E. Tomaszewski, R. Yacoub, F. Chen, and P. Sarder, “Unsupervised labeling of glomerular boundaries using Gabor filters and statistical testing in renal histology,” Journal of Medical Imaging – SPIE, vol. 4, no. 2, pp. 021102:1–13, Feb. 2017.