Predicting Clinical Endpoints and Visual Changes with Quality-weighted Tissue-based Renal Histological Features

Overview

Journal Front Transplant

Specialty General Surgery

Date 2024 Jul 12

PMID 38993786

Authors

Ka Ho Tam

Maria F Soares

Jesper Kers

Edward J Sharples

Rutger J Ploeg

Maria Kaisar

Jens Rittscher

Affiliations

Soon will be listed here.

Abstract

Two common obstacles limiting the performance of data-driven algorithms in digital histopathology classification tasks are the lack of expert annotations and the narrow diversity of datasets. Multi-instance learning (MIL) can address the former challenge for the analysis of whole slide images (WSI), but performance is often inferior to full supervision. We show that the inclusion of weak annotations can significantly enhance the effectiveness of MIL while keeping the approach scalable. An analysis framework was developed to process periodic acid-Schiff (PAS) and Sirius Red (SR) slides of renal biopsies. The workflow segments tissues into coarse tissue classes. Handcrafted and deep features were extracted from these tissues and combined using a soft attention model to predict several slide-level labels: delayed graft function (DGF), acute tubular injury (ATI), and Remuzzi grade components. A tissue segmentation quality metric was also developed to reduce the adverse impact of poorly segmented instances. The soft attention model was trained using 5-fold cross-validation on a mixed dataset and tested on the QUOD dataset containing PAS and SR biopsies. The average ROC-AUC over different prediction tasks was found to be , significantly higher than using only ResNet50 ( ), only handcrafted features ( ), and the baseline ( ) of state-of-the-art performance. In conjunction with soft attention, weighting tissues by segmentation quality has led to further improvement . Using an intuitive visualisation scheme, we show that our approach may also be used to support clinical decision making as it allows pinpointing individual tissues relevant to the predictions.

References

Lu M, Williamson D, Chen T, Chen R, Barbieri M, Mahmood F . Data-efficient and weakly supervised computational pathology on whole-slide images. Nat Biomed Eng. 2021; 5(6):555-570. PMC: 8711640. DOI: 10.1038/s41551-020-00682-w. View

Campanella G, Hanna M, Geneslaw L, Miraflor A, Silva V, Busam K . Clinical-grade computational pathology using weakly supervised deep learning on whole slide images. Nat Med. 2019; 25(8):1301-1309. PMC: 7418463. DOI: 10.1038/s41591-019-0508-1. View

Racusen L, Solez K, Colvin R, Bonsib S, Castro M, CAVALLO T . The Banff 97 working classification of renal allograft pathology. Kidney Int. 1999; 55(2):713-23. DOI: 10.1046/j.1523-1755.1999.00299.x. View

Bruna J, Mallat S . Invariant scattering convolution networks. IEEE Trans Pattern Anal Mach Intell. 2013; 35(8):1872-86. DOI: 10.1109/TPAMI.2012.230. View

Weissenbacher A, Faro L, Boubriak O, Soares M, Roberts I, Hunter J . Twenty-four-hour normothermic perfusion of discarded human kidneys with urine recirculation. Am J Transplant. 2018; 19(1):178-192. PMC: 6491986. DOI: 10.1111/ajt.14932. View

Iizuka O, Kanavati F, Kato K, Rambeau M, Arihiro K, Tsuneki M . Deep Learning Models for Histopathological Classification of Gastric and Colonic Epithelial Tumours. Sci Rep. 2020; 10(1):1504. PMC: 6992793. DOI: 10.1038/s41598-020-58467-9. View

Li X, Davis R, Xu Y, Wang Z, Souma N, Sotolongo G . Deep learning segmentation of glomeruli on kidney donor frozen sections. J Med Imaging (Bellingham). 2021; 8(6):067501. PMC: 8685284. DOI: 10.1117/1.JMI.8.6.067501. View

Solez K, Axelsen R, Benediktsson H, Burdick J, Cohen A, Colvin R . International standardization of criteria for the histologic diagnosis of renal allograft rejection: the Banff working classification of kidney transplant pathology. Kidney Int. 1993; 44(2):411-22. DOI: 10.1038/ki.1993.259. View

Rolak S, Djamali A, Mandelbrot D, Muth B, Jorgenson M, Zhong W . Outcomes of Delayed Graft Function in Kidney Transplant Recipients Stratified by Histologic Biopsy Findings. Transplant Proc. 2021; 53(5):1462-1469. DOI: 10.1016/j.transproceed.2021.01.012. View

10.

Hermsen M, de Bel T, den Boer M, Steenbergen E, Kers J, Florquin S . Deep Learning-Based Histopathologic Assessment of Kidney Tissue. J Am Soc Nephrol. 2019; 30(10):1968-1979. PMC: 6779356. DOI: 10.1681/ASN.2019020144. View

11.

Kers J, Peters-Sengers H, Heemskerk M, Berger S, Betjes M, van Zuilen A . Prediction models for delayed graft function: external validation on The Dutch Prospective Renal Transplantation Registry. Nephrol Dial Transplant. 2018; 33(7):1259-1268. DOI: 10.1093/ndt/gfy019. View

12.

Li B, Li Y, Eliceiri K . Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learning. Conf Comput Vis Pattern Recognit Workshops. 2022; 2021:14318-14328. PMC: 8765709. DOI: 10.1109/CVPR46437.2021.01409. View

13.

Kers J, Bulow R, Klinkhammer B, Breimer G, Fontana F, Abiola A . Deep learning-based classification of kidney transplant pathology: a retrospective, multicentre, proof-of-concept study. Lancet Digit Health. 2021; 4(1):e18-e26. DOI: 10.1016/S2589-7500(21)00211-9. View

14.

Yi Z, Salem F, Menon M, Keung K, Xi C, Hultin S . Deep learning identified pathological abnormalities predictive of graft loss in kidney transplant biopsies. Kidney Int. 2021; 101(2):288-298. PMC: 10285669. DOI: 10.1016/j.kint.2021.09.028. View

15.

Remuzzi G, Cravedi P, Perna A, Dimitrov B, Turturro M, Locatelli G . Long-term outcome of renal transplantation from older donors. N Engl J Med. 2006; 354(4):343-52. DOI: 10.1056/NEJMoa052891. View

16.

Yarlagadda S, Coca S, Garg A, Doshi M, Poggio E, Marcus R . Marked variation in the definition and diagnosis of delayed graft function: a systematic review. Nephrol Dial Transplant. 2008; 23(9):2995-3003. PMC: 2727302. DOI: 10.1093/ndt/gfn158. View

17.

Marsh J, Matlock M, Kudose S, Liu T, Stappenbeck T, Gaut J . Deep Learning Global Glomerulosclerosis in Transplant Kidney Frozen Sections. IEEE Trans Med Imaging. 2018; 37(12):2718-2728. PMC: 6296264. DOI: 10.1109/TMI.2018.2851150. View

18.

Farris A, Adams C, Brousaides N, Della Pelle P, Collins A, Moradi E . Morphometric and visual evaluation of fibrosis in renal biopsies. J Am Soc Nephrol. 2010; 22(1):176-86. PMC: 3014046. DOI: 10.1681/ASN.2009091005. View

19.

Grimm P, Nickerson P, Gough J, McKenna R, Stern E, Jeffery J . Computerized image analysis of Sirius Red-stained renal allograft biopsies as a surrogate marker to predict long-term allograft function. J Am Soc Nephrol. 2003; 14(6):1662-8. DOI: 10.1097/01.asn.0000066143.02832.5e. View

20.

Pieters T, Falke L, Nguyen T, Verhaar M, Florquin S, Bemelman F . Histological characteristics of Acute Tubular Injury during Delayed Graft Function predict renal function after renal transplantation. Physiol Rep. 2019; 7(5):e14000. PMC: 6395310. DOI: 10.14814/phy2.14000. View