Supervised Computational Detection of Glomeruli in Renal Biopsies

In this work, we have developed a computational pipeline to detect glomeruli from whole-slide images (WSIs) of renal tissue sections marked with histological stains. Multi-radial color local binary pattern (mrcLBP) features extracted from glomerular and non-glomerular regions are used to train a support vector machine (SVM), deployed in tandem with a deep convolutional neural network trained for glomerular recognition. Using 5 Intel(R) Core(TM) i7-4790 CPUs with 40 GB RAM, the method takes ~2 min to extract glomeruli from a biopsy WSI (~106 pixels).

Examples and results from glomerular detection in human WSIs using mrcLBP.
Examples and results from glomerular detection in human WSIs using mrcLBP a) Examples of successful application of mrcLBP to human renal WSIs, with detected glomeruli (green boxes). Top, renal section. Bottom, needle biopsy.  PAS stain.  b) Results of detection of glomeruli in human tissue samples.  Left group, non-diabetic nephropathy (DN) patients, from 9 slides containing 1088 glomeruli. Right group, result from DN patients, from 5 slides containing 99 glomeruli.

Select References

O. Simon, R. Yacoub, S. Jain, J. E. Tomaszewski, and P. Sarder, “Multi-radial LBP features as a tool for rapid glomerular detection and assessment in whole slide histopathology images,” Scientific Reports – Nature, vol. 8, pp. 2032:1–11, Feb. 2018.

Underline indicates corresponding author.
* indicates equal contribution.
indicates Dr. Sarder’s faculty trainees.
§ indicates post doctoral associates.
indicates Dr. Sarder’s graduate students.
indicates Dr. Sarder’s undergraduate students.