Dataset Information

Automated Computational Detection of Interstitial Fibrosis, Tubular Atrophy, and Glomerulosclerosis.

ABSTRACT: Interstitial fibrosis, tubular atrophy (IFTA), and glomerulosclerosis are indicators of irrecoverable kidney injury. Modern machine learning (ML) tools have enabled robust, automated identification of image structures that can be comparable with analysis by human experts. ML algorithms were developed and tested for the ability to replicate the detection and quantification of IFTA and glomerulosclerosis that renal pathologists perform. A renal pathologist annotated renal biopsy specimens from 116 whole-slide images (WSIs) for IFTA and glomerulosclerosis. A total of 79 WSIs were used for training different configurations of a convolutional neural network (CNN), and 17 and 20 WSIs were used as internal and external testing cases, respectively. The best model was compared against the input of four renal pathologists on 20 new testing slides. Further, for 87 testing biopsy specimens, IFTA and glomerulosclerosis measurements made by pathologists and the CNN were correlated to patient outcome using classic statistical tools. The best average performance across all image classes came from a DeepLab version 2 network trained at 40× magnification. IFTA and glomerulosclerosis percentages derived from this CNN achieved high levels of agreement with four renal pathologists. The pathologist- and CNN-based analyses of IFTA and glomerulosclerosis showed statistically significant and equivalent correlation with all patient-outcome variables. ML algorithms can be trained to replicate the IFTA and glomerulosclerosis assessment performed by renal pathologists. This suggests computational methods may be able to provide a standardized approach to evaluate the extent of chronic kidney injury in situations in which renal-pathologist time is restricted or unavailable.

SUBMITTER: Ginley B

PROVIDER: S-EPMC8017538 | biostudies-literature | 2021 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Automated Computational Detection of Interstitial Fibrosis, Tubular Atrophy, and Glomerulosclerosis.

Ginley Brandon B Jen Kuang-Yu KY Han Seung Seok SS Rodrigues Luís L Rodrigues Luís L Jain Sanjay S Fogo Agnes B AB Zuckerman Jonathan J Walavalkar Vighnesh V Miecznikowski Jeffrey C JC Wen Yumeng Y Yen Felicia F Yun Donghwan D Moon Kyung Chul KC Rosenberg Avi A Rosenberg Avi A Parikh Chirag C Sarder Pinaki P

Journal of the American Society of Nephrology : JASN 20210223 4

<h4>Background</h4>Interstitial fibrosis, tubular atrophy (IFTA), and glomerulosclerosis are indicators of irrecoverable kidney injury. Modern machine learning (ML) tools have enabled robust, automated identification of image structures that can be comparable with analysis by human experts. ML algorithms were developed and tested for the ability to replicate the detection and quantification of IFTA and glomerulosclerosis that renal pathologists perform.<h4>Methods</h4>A renal pathologist annotat ...[more]

PMID: 33622976

Similar Datasets

Project description:BackgroundThis study examines the effect of interstitial inflammation and interstitial fibrosis and tubular atrophy on renal survival in lupus nephritis.MethodsBaseline characteristics, initial (n = 301) and repeat biopsies (n = 94) and clinical outcomes for patients with biopsy-proven lupus nephritis from 1998 to 2014 were retrospectively collected from the medical record. Clinical and morphologic variables were evaluated using a Cox proportional hazards model and multiple imputation to address missing data. Renal survival was defined as the time from initial biopsy to end-stage renal disease [estimated glomerular filtration rate (eGFR) <15 mL/min/1.73 m2], dialysis or transplant.ResultsA total of 218 patients had follow-up and Class IV had worse renal survival, especially in patients with active and chronic glomerular lesions {relative to non-IV; Class IV-A: hazard ratio [HR] 0.92 [95% confidence interval (CI) 0.41-2.04], Class IV-AC: HR 5.02 [95% CI 2.70-9.36]}. Interstitial inflammation grade [relative to interstitial inflammation <5%; interstitial inflammation 5-25%: HR 2.36 (95% CI 1.13-4.91), interstitial inflammation 25-50%: HR 3.84 (95% CI 1.53-9.62), interstitial inflammation >50%: HR 7.67 (95% CI 3.75-15.67)] and increased interstitial fibrosis and tubular atrophy (IFTA) category [relative to IFTA <5%; IFTA 5-25%: HR 3.93 (95% CI 1.58-9.75), IFTA 25-50%: HR 4.01 (95% CI 1.37-11.70), IFTA >50%: HR 13.99 (95% CI 4.91-39.83)] predicted worse renal survival among all patients and those with Class IV on initial and repeat biopsy (n = 94) in a dose-dependent manner. Interstitial inflammation grade and IFTA category were significant predictors of renal survival in a multivariable model adjusted for age, gender, race, ethnicity and serum creatinine.ConclusionsInterstitial inflammation and IFTA independently affect renal survival and grading these lesions stratifies risk within the International Society of Nephrology and Renal Pathology Society classification of lupus nephritis.

Project description:BackgroundIntrarenal complement activation has been implicated in the pathogenesis of tubulointerstitial fibrosis in lupus nephritis (LN) based on prior animal studies. The assembly of the membrane attack complex (MAC) by complement C5b to C9 on the cell membrane leads to cytotoxic pores and cell lysis, while CD59 inhibits MAC formation by preventing C9 from joining the complex. We hypothesize that complement activation and imbalance between complement activation and inhibition, as defined by increased production of individual complement components and uncontrolled MAC activation relative to CD59 inhibition, are associated with interstitial fibrosis and tubular atrophy (IFTA) in LN and correlate with the key mediators of kidney fibrosis- transforming growth factor receptors beta (TGFRβ), platelet-derived growth factor beta (PDGFβ) and platelet-derived growth factor receptor beta (PDGFRβ).MethodsWe included urine samples from 46 adults and pediatric biopsy-proven lupus nephritis patients who underwent clinically indicated kidney biopsies between 2010 and 2019. We compared individual urinary complement components and the urinary C9-to-CD59 ratio between LN patients with moderate/severe IFTA and none/mild IFTA. IFTA was defined as none/mild (<25% of interstitium affected) versus moderate/severe (≥ 25% of interstitium affected). Proteomics analysis was performed using mass spectrometry (Orbitrap Fusion Lumos, Thermo Scientific) and processed by the Proteome Discoverer. Urinary complement proteins enriched in LN patients with moderate/severe IFTA were correlated with serum creatinine, TGFβR1, TGFβR2, PDGFβ, and PDGFRβ.ResultsOf the 46 LN patients included in the study, 41 (89.1%) were women, 20 (43.5%) self-identified as Hispanic or Latino, and 26 (56.5%) self-identified as Black or African American. Ten of the 46 (21.7%) LN patients had moderate/severe IFTA on kidney biopsy. LN patients with moderate/severe IFTA had an increased urinary C9-to-CD59 ratio [median 0.91 (0.83-1.05) vs 0.81 (0.76-0.91), p=0.01]. Urinary C3 and CFI levels in LN patients with moderate/severe IFTA were higher compared to those with none/mild IFTA [C3 median (IQR) 24.4(23.5-25.5) vs. 20.2 (18.5-22.2), p= 0.02], [CFI medium (IQR) 28.8 (21.8-30.6) vs. 20.4 (18.5-22.9), p=0.01]. Complement C9, CD59, C3 and CFI correlated with TGFβR1, PDGFβ, and PDGFRβ, while C9, CD59 and C3 correlated with TGFβR2.ConclusionThis study is one of the first to compare the urinary complement profile in LN patients with moderate/severe IFTA and none/mild IFTA in human tissues. This study identified C3, CFI, and C9-to-CD59 ratio as potential markers of tubulointerstitial fibrosis in LN.

Project description:Background: Interstitial fibrosis and tubular atrophy (IFTA) are the histopathological manifestations of chronic kidney disease (CKD) and one of the causes of long-term renal loss in transplanted kidneys. Necroptosis as a type of programmed death plays an important role in the development of IFTA, and in the late functional decline and even loss of grafts. In this study, 13 machine learning algorithms were used to construct IFTA diagnostic models based on necroptosis-related genes. Methods: We screened all 162 "kidney transplant"-related cohorts in the GEO database and obtained five data sets (training sets: GSE98320 and GSE76882, validation sets: GSE22459 and GSE53605, and survival set: GSE21374). The training set was constructed after removing batch effects of GSE98320 and GSE76882 by using the SVA package. The differentially expressed gene (DEG) analysis was used to identify necroptosis-related DEGs. A total of 13 machine learning algorithms-LASSO, Ridge, Enet, Stepglm, SVM, glmboost, LDA, plsRglm, random forest, GBM, XGBoost, Naive Bayes, and ANNs-were used to construct 114 IFTA diagnostic models, and the optimal models were screened by the AUC values. Post-transplantation patients were then grouped using consensus clustering, and the different subgroups were further explored using PCA, Kaplan-Meier (KM) survival analysis, functional enrichment analysis, CIBERSOFT, and single-sample Gene Set Enrichment Analysis. Results: A total of 55 necroptosis-related DEGs were identified by taking the intersection of the DEGs and necroptosis-related gene sets. Stepglm[both]+RF is the optimal model with an average AUC of 0.822. A total of four molecular subgroups of renal transplantation patients were obtained by clustering, and significant upregulation of fibrosis-related pathways and upregulation of immune response-related pathways were found in the C4 group, which had poor prognosis. Conclusion: Based on the combination of the 13 machine learning algorithms, we developed 114 IFTA classification models. Furthermore, we tested the top model using two independent data sets from GEO.

Dataset Information

Automated Computational Detection of Interstitial Fibrosis, Tubular Atrophy, and Glomerulosclerosis.

Publications

Automated Computational Detection of Interstitial Fibrosis, Tubular Atrophy, and Glomerulosclerosis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets