Machine learning-supported interpretation of kidney graft elementary lesions in combination with clinical data.
biopsy
classification systems: Banff classification
clinical research / practice
kidney transplantation / nephrology
rejection: T cell mediated (TCMR)
rejection: antibody-mediated (ABMR)
Journal
American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons
ISSN: 1600-6143
Titre abrégé: Am J Transplant
Pays: United States
ID NLM: 100968638
Informations de publication
Date de publication:
12 2022
12 2022
Historique:
revised:
22
08
2022
received:
08
02
2022
accepted:
25
08
2022
pubmed:
6
9
2022
medline:
6
12
2022
entrez:
5
9
2022
Statut:
ppublish
Résumé
Interpretation of kidney graft biopsies using the Banff classification is still heterogeneous. In this study, extreme gradient boosting classifiers learned from two large training datasets (n = 631 and 304 cases) where the "reference diagnoses" were not strictly defined following the Banff rules but from central reading by expert pathologists and further interpreted consensually by experienced transplant nephrologists, in light of the clinical context. In three external validation datasets (n = 3744, 589, and 360), the classifiers yielded a mean ROC curve AUC (95%CI) of: 0.97 (0.92-1.00), 0.97 (0.96-0.97), and 0.95 (0.93-0.97) for antibody-mediated rejection (ABMR); 0.94 (0.91-0.96), 0.94 (0.92-0.95), and 0.91 (0.88-0.95) for T cell-mediated rejection; >0.96 (0.90-1.00) with all three for interstitial fibrosis-tubular atrophy. We also developed a classifier to discriminate active and chronic active ABMR with 95% accuracy. In conclusion, we built highly sensitive and specific artificial intelligence classifiers able to interpret kidney graft scoring together with a few clinical data and automatically diagnose rejection, with excellent concordance with the Banff rules and reference diagnoses made by a group of experts. Some discrepancies may point toward possible improvements that could be made to the Banff classification.
Identifiants
pubmed: 36062389
doi: 10.1111/ajt.17192
pii: S1600-6135(23)00034-5
doi:
Substances chimiques
Isoantibodies
0
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
2821-2833Commentaires et corrections
Type : CommentIn
Informations de copyright
© 2022 The American Society of Transplantation and the American Society of Transplant Surgeons.
Références
Loupy A, Haas M, Roufosse C, et al. The Banff 2019 kidney meeting report (I): updates on and clarification of criteria for T cell- and antibody-mediated rejection. Am J Transplant. 2020;20(9):2318-2331. doi:10.1111/ajt.15898
Marcussen N, Olsen TS, Benediktsson H, Racusen L, Solez K. Reproducibility of the Banff classification of renal allograft pathology. Inter- and intraobserver variation. Transplantation. 1995;60(10):1083-1089.
Furness PN, Taub N. Convergence of European renal transplant pathology assessment procedures (CERTPAP) project. International variation in the interpretation of renal transplant biopsies: report of the CERTPAP project. Kidney Int. 2001;60(5):1998-2012. doi:10.1046/j.1523-1755.2001.00030.x
Furness PN, Taub N, Assmann KJM, et al. International variation in histologic grading is large, and persistent feedback does not improve reproducibility. Am J Surg Pathol. 2003;27(6):805-810.
Schinstock CA, Sapir-Pichhadze R, Naesens M, et al. Banff survey on antibody mediated rejection clinical practices in kidney transplantation: diagnostic misinterpretation has potential therapeutic implications. Am J Transplant. 2019;19(1):123-131. doi:10.1111/ajt.14979
Vaulet T, Divard G, Thaunat O, et al. Data-driven derivation and validation of novel phenotypes for acute kidney transplant rejection using semi-supervised clustering. J Am Soc Nephrol JASN. 2021;32(5):1084-1096. doi:10.1681/ASN.2020101418
Marx D, Metzger J, Olagne J, et al. Proteomics in kidney allograft transplantation-application of molecular pathway analysis for kidney allograft disease phenotypic biomarker selection. Proteomics Clin Appl. 2019;13(2):e1800091. doi:10.1002/prca.201800091
Rabant M, Amrouche L, Lebreton X, et al. Urinary C-X-C motif chemokine 10 independently improves the noninvasive diagnosis of antibody-mediated kidney allograft rejection. J Am Soc Nephrol JASN. 2015;26(11):2840-2851. doi:10.1681/ASN.2014080797
Haas M, Sis B, Racusen LC, et al. Banff 2013 meeting report: inclusion of c4d-negative antibody-mediated rejection and antibody-associated arterial lesions. Am J Transplant. 2014;14(2):272-283. doi:10.1111/ajt.12590
Haas M. The revised (2013) Banff classification for antibody-mediated rejection of renal allografts: update, difficulties, and future considerations. Am J Transplant. 2016;16(5):1352-1357. doi:10.1111/ajt.13661
Chen T, Guestrin C. XGBoost: a scalable tree boosting system. ArXiv160302754 cs. Published online June 10, 2016. 10.1145/2939672.2939785
XGBoost. Accessed April 21, 2021. https://kaggle.com/dansbecker/xgboost
Davis J, Goadrich M. The relationship between precision-recall and ROC curves. In: Proceedings of the 23rd International Conference on Machine Learning. ICML ‘06. Association for Computing Machinery; 2006:233-240. 10.1145/1143844.1143874
Saito T, Rehmsmeier M. The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PloS One. 2015;10(3):e0118432. doi:10.1371/journal.pone.0118432
Greenwell B. Fastshap: Fast Approximate Shapley Values. 2020. Accessed October 14, 2021. https://CRAN.R-project.org/package=fastshap
Cannesson M. The “grey zone” or how to avoid the binary constraint of decision-making. Can J Anaesth. 2015;62(11):1139-1142. doi:10.1007/s12630-015-0465-1
Hastie T, Tibshirani R, Friedman J. The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Second Edition. 2nd ed. Springer-Verlag; 2009. doi:10.1007/978-0-387-84858-7
Wohlfahrtova M, Hruba P, Klema J, et al. Early isolated V-lesion may not truly represent rejection of the kidney allograft. Clin Sci. 2018;132(20):2269-2284. doi:10.1042/CS20180745
Senev A, Coemans M, Lerut E, et al. Histological picture of antibody-mediated rejection without donor-specific anti-HLA antibodies: clinical presentation and implications for outcome. Am J Transplant. 2019;19(3):763-780. doi:10.1111/ajt.15074
Smith B, Cornell LD, Smith M, et al. A method to reduce variability in scoring anti-body mediated rejection in renal allografts: implications for clinical trials. Transpl Int. 2019;32(2):173-183. doi:10.1111/tri.13340
Gough J, Rush D, Jeffery J, et al. Reproducibility of the Banff schema in reporting protocol biopsies of stable renal allografts. Nephrol Dial Transplant. 2002;17(6):1081-1084. doi:10.1093/ndt/17.6.1081
Veronese FV, Manfro RC, Roman FR, et al. Reproducibility of the Banff classification in subclinical kidney transplant rejection. Clin Transplant. 2005;19(4):518-521. doi:10.1111/j.1399-0012.2005.00377.x
Gadermayr M, Dombrowski AK, Klinkhammer BM, Boor P, Merhof D. CNN cascades for segmenting sparse objects in gigapixel whole slide images. Comput Med Imaging Graph. 2019;71:40-48. doi:10.1016/j.compmedimag.2018.11.002
Pedraza A, Gallego J, Lopez S, Gonzalez L, Laurinavicius A, Bueno G. Glomerulus classification with convolutional neural networks. In: Valdés Hernández M, González-Castro V, eds. Medical Image Understanding and Analysis. Communications in Computer and Information Science. Springer International Publishing; 2017:839-849. 10.1007/978-3-319-60964-5_73
Bukowy JD, Dayton A, Cloutier D, et al. Region-based convolutional neural nets for localization of glomeruli in trichrome-stained whole kidney sections. J Am Soc Nephrol. 2018;29(8):2081-2088. doi:10.1681/ASN.2017111210
Hermsen M, de Bel T, den Boer M, et al. Deep learning-based histopathologic assessment of kidney tissue. J Am Soc Nephrol. 2019;30(10):1968-1979. doi:10.1681/ASN.2019020144
Arthurs C, Roufosse C. Forging the tools for a computer-aided workflow in transplant pathology. Lancet Digit Health. 2022;4(1):e2-e3. doi:10.1016/S2589-7500(21)00254-5