Abstract:Interstitial fibrosis / tubular atrophy (IFTA) is a common, irreversible and progressive form of chronic allograft injury, and it is considered a critical predictor of kidney allograft outcomes. Inflammation, both microvascular and interstitial, is on the contrary regarded as a reversible form of graft injury. Since treatments for rejection and other causes of graft dysfunction bear substantial toxicity and could have limited efficacy, the extent of irreversible graft scarring is a crucial information for the clinician, to evaluate risks and benefits of specific therapies. The diagnosis of kidney graft pathology is acquired through graft biopsy, which is an invasive procedure and can be subjected to sampling bias. Magnetic resonance imaging (MRI), especially with functional techniques, has emerged as a possibility for non-invasive estimation of tissue fibrosis; nevertheless, functional MRI is not widely available. Texture analysis MRI (TA-MRI) is a radiomic technique that provides a quantitative assessment of tissue heterogeneity from standard MRI images, generating features that can be fitted into a machine-learning model to assess their ability to predict clinical or histological parameters. Single-center cross-sectional observational cohort study enrolling kidney transplant recipients who underwent graft biopsy and graft MRI imaging within 6 months from biopsy, both on clinical indication, at the “Azienda Ospedaliero-Universitaria di Modena”, Italy. The study was approved by the local Ethical Committee (AOU0010167/20). The primary outcome was to identify the best TA-MRI features subset for estimation of IFTA > 50% in graft biopsy. Secondary outcomes were estimation of: IFTA > 25%, presence of total inflammation (ti) and microvascular inflammation (glomerulitis + peritubular capillaritis [g+ptc]). Graft biopsy was reported according to Banff 2017 system. Radiomic analysis was performed on axial T2 pre-contrast and T1 fat-suppressed post-contrast sequences. The whole renal parenchyma (PAR) was segmented and labelled on T2 and T1, renal cortex (COR) only on T2. After imaging preprocessing, PyRadiomics was used to extract radiomic features. After removal of shape features, 93 features were included and reduced using LASSO regression to produce radiomic signatures. These were introduced in Machine Learning (ML) models to test the association with outcomes. Results are reported as AUC and a value of sensitivity and specificity. Sixty patients were included in the study, and 67 graft biopsy – graft MRI pairs were available for analysis. Demographic and clinical characteristics of enrolled patients are depicted in table 1; histological diagnosis and main Banff histological parameters from graft biopsies in table 2. Among ML models, three showed an acceptable performance. T2 COR “firstorder_minimum/firstorder_range/glrlm_run_entropy” for IFTA>50% (AUC=0.77, sensitivity=73%, specificity=71%), T1 PAR “firstorder_energy” for IFTA>25% (AUC=0.71, sensitivity=74%, specificity=51%), T1 PAR “firstorder_energy/gldm_small_dependence_low_gray_level_emphasis” for g+ptc >0 (AUC=0.74, sensitivity= 78%, specificity=68%); see figures 1–3. No acceptable prediction was detected for ti >0. Our study shows that TA-MRI feature signatures can predict the degree of IFTA in graft biopsies, with an acceptable diagnostic performance. These results suggest to further investigating TA-MRI from standard MRI sequences as potential tool to assess graft chronic parenchymal injury. Moreover, since graft biopsy results can be jeopardized by limited sample size, we hypothesize that evaluation of IFTA through TA-MRI could provide more comprehensive information regarding the whole parenchyma. To test this hypothesis, we are currently evaluating the association of TA-MRI radiomic features and baseline eGFR and eGFR variation over time.

Clinical Relevance of Computationally Derived Tubular Features: Spatial Relationships and the Development of Tubulointerstitial Scarring in MCD/FSGS

Computational pathology model to assess acute and chronic transformations of the tubulointerstitial compartment in renal allograft biopsies

Quantitative assessment of inflammatory infiltrates in kidney transplant biopsies using multiplex tyramide signal amplification and deep learning

Deep-Learning-Driven Quantification of Interstitial Fibrosis in Digitized Kidney Biopsies

Development and Validation of a Deep Learning Model to Quantify Interstitial Fibrosis and Tubular Atrophy From Kidney Ultrasonography Images

Neural Network Segmentation of Interstitial Fibrosis, Tubular Atrophy, and Glomerulosclerosis in Renal Biopsies

Recognition of intraglomerular histological features with deep learning in protocol transplant biopsies and their association with kidney function and prognosis

Exploring the Predictive Value of Additional Peritumoral Regions Based on Deep Learning and Radiomics: A Multicenter Study.

CT-based conventional radiomics and quantification of intratumoral heterogeneity for predicting benign and malignant renal lesions

Digital pathology assessment of kidney glomerular filtration barrier ultrastructure in an animal model of podocytopathy

Automated evaluation with deep learning of total interstitial inflammation and peritubular capillaritis on kidney biopsies

Computational features of tumor-infiltrating lymphocyte architecture of residual disease after chemotherapy on H&E images as prognostic of overall and disease-free survival for triple-negative breast cancer.

Correlation of glomerular histomorphometry changes with spatially resolved transcriptomic profiles in diabetic nephropathy

Artificial intelligence assists identification and pathologic classification of glomerular lesions in patients with diabetic nephropathy

Integration of spatial multiplexed protein imaging and transcriptomics in the human kidney tracks the regenerative potential timeline of proximal tubules

Abstract P2-11-16: Computerized Measurements of Nuclear Morphology Features, Mitosis Rate, and Tubule Formation from H&E Images Predicts Disease-Free Survival in Patients with HR+ & LN+ Invasive Breast Cancer from SWOG S8814

Construction of predictive model of interstitial fibrosis and tubular atrophy after kidney transplantation with machine learning algorithms

Intestinal fibrosis classification in patients with Crohn’s disease using CT enterography–based deep learning: comparisons with radiomics and radiologists

Differentiation of low and high grade renal cell carcinoma on routine MRI with an externally validated automatic machine learning algorithm

FC 120MAGNETIC RESONANCE IMAGING TEXTURE ANALYSIS PREDICTS INTERSTITIAL FIBROSIS / TUBULAR ATROPHY IN TRANSPLANTED KIDNEYS: A SINGLE CENTER CROSS-SECTIONAL STUDY

Identification of glomerular lesions and intrinsic glomerular cell types in kidney diseases via deep learning