Deep Learning Global Glomerulosclerosis in Transplant Kidney Frozen Sections
Preprint
Abstract
Transplantable kidneys are in very limited supply. Accurate viability assessment prior to transplantation could minimize organ discard. Rapid and accurate evaluation of intra-operative donor kidney biopsies is essential for determining which kidneys are eligible for transplantation. The criteria for accepting or rejecting donor kidneys relies heavily on pathologist determination of the percent of glomeruli (determined from a frozen section) that are normal and sclerotic. This percentage is a critical measurement that correlates with transplant outcome. Inter- and intra-observer variability in donor biopsy evaluation is, however, significant. An automated method for determination of percent global glomerulosclerosis could prove useful in decreasing evaluation variability, increasing throughput, and easing the burden on pathologists. Here, we describe the development of a deep learning model that identifies and classifies non-sclerosed and sclerosed glomeruli in whole-slide images of donor kidney frozen section biopsies. This model extends a convolutional neural network (CNN) pre-trained on a large database of digital images. The extended model, when trained on just 48 whole slide images, exhibits slide-level evaluation performance on par with expert renal pathologists. The model substantially outperforms a model trained on image patches of isolated glomeruli. Encouragingly, the model’s performance is robust to slide preparation artifacts associated with frozen section preparation. As the first model reported that identifies and classifies normal and sclerotic glomeruli in frozen kidney sections, and thus the first model reported in the literature relevant to kidney transplantation, it may become an essential part of donor kidney biopsy evaluation in the clinical setting.Keywords
All Related Versions
- Published version: IEEE Transactions on Medical Imaging, 37 (12), 2718.
This publication has 36 references indexed in Scilit:
- The reproducibility and predictive value on outcome of renal biopsies from expanded criteria donorsKidney International, 2014
- Automated image analysis of a glomerular injury marker desmin in spontaneously diabetic Torii rats treated with losartanJournal of Endocrinology, 2014
- Fiji: an open-source platform for biological-image analysisNature Methods, 2012
- Computer-assisted imaging algorithms facilitate histomorphometric quantification of kidney damage in rodent renal failure modelsJournal of Pathology Informatics, 2012
- Graft quality assessment in kidney transplantation: not an exact science yet!Current Opinion in Organ Transplantation, 2011
- A Simple Clinico-Histopathological Composite Scoring System Is Highly Predictive of Graft Outcomes in Marginal DonorsAmerican Journal of Transplantation, 2008
- The Maryland Aggregate Pathology Index: A Deceased Donor Kidney Biopsy Scoring System for Predicting Graft FailureAmerican Journal of Transplantation, 2008
- Long-Term Outcome of Renal Transplantation from Older DonorsNew England Journal of Medicine, 2006
- International variation in the interpretation of renal transplant biopsies: Report of the CERTPAP Project11A complete listing of the participants who all contributed equally to the work of the CERTPAP Project includes Peter N. Furness (Leicester, UK); Nicholas Taub (Leicester, United Kingdom); Karel J.M. Assmann (Nijmegen, The Netherlands); Giovanni Banfi (Milan, Italy); John N. Botelis (Athens, Greece); Marta Carrera (Barcelona, Spain); Jean-Pierre Cosyns (Brussels, Belgium); Anthony M. Dorman (Dublin, Ireland); Dominique Droz (Paris, France); Claire M. Hill (Belfast, N. Ireland); Bela Iványi (Kossuth, Hungary); Silke Kapper (Mannheim, Germany); Erik N. Larsson (Uppsala, Sweden); Aryvdas Laurinavicius (Vilius, Lithuania); Niels Marcussen (Aachus, Denmark); Anna Paula Martins (Lisbon, Portugal); Michael J. Mihatsch (Basel, Switzerland); Lydia Nakopoulou (Athens, Greece); Volker Nickeleit (Basel, Switzerland); L-H Nöel (Paris, France); Timo Paavonen (Helsinki, Finland); Agnieszua K. Perkowska (Warsaw, Poland); Heinz Regele (Vienna, Austria); Rafail Rosenthal (Riga, Latvia); Pavel Rossmann (Prague, Czech Republic); Wotgech A. Rowinski (Warsaw, Poland); Daniel Seron (Barcelona, Spain); Stale Sund (Oslo, Norway); Eero I. Taskinen (Helsinki, Finland); Tatjana Tihomirova (Riga, Latvia); and Rudiger Waldherr (Mannheim, Germany).Kidney International, 2001
- Detecting salient blob-like image structures and their scales with a scale-space primal sketch: A method for focus-of-attentionInternational Journal of Computer Vision, 1993