BackgroundThe development of deep neural networks is facilitating more advanced digital analysis of histopathologic images. We trained a convolutional neural network for multiclass segmentation of digitized kidney tissue sections stained with periodic acid–Schiff (PAS).MethodsWe trained the network using multiclass annotations from 40 whole-slide images of stained kidney transplant biopsies and applied it to four independent data sets. We assessed multiclass segmentation performance by calculating Dice coefficients for ten tissue classes on ten transplant biopsies from the Radboud University Medical Center in Nijmegen, The Netherlands, and on ten transplant biopsies from an external center for validation. We also fully segmented 15 nephrectomy samples and calculated the network’s glomerular detection rates and compared network-based measures with visually scored histologic components (Banff classification) in 82 kidney transplant biopsies.ResultsThe weighted mean Dice coefficients of all classes were 0.80 and 0.84 in ten kidney transplant biopsies from the Radboud center and the external center, respectively. The best segmented class was “glomeruli” in both data sets (Dice coefficients, 0.95 and 0.94, respectively), followed by “tubuli combined” and “interstitium.” The network detected 92.7% of all glomeruli in nephrectomy samples, with 10.4% false positives. In whole transplant biopsies, the mean intraclass correlation coefficient for glomerular counting performed by pathologists versus the network was 0.94. We found significant correlations between visually scored histologic components and network-based measures.ConclusionsThis study presents the first convolutional neural network for multiclass segmentation of PAS-stained nephrectomy samples and transplant biopsies. Our network may have utility for quantitative studies involving kidney histopathology across centers and provide opportunities for deep learning applications in routine diagnostics.

No abstract

No abstract

he prospect of improved clinical outcomes and more efficient health systems has fueled a rapid rise in the development and evaluation of AI systems over the last decade. Because most AI systems within healthcare are complex interventions designed as clinical decision support systems, rather than autonomous agents, the interactions among the AI systems, their users and the implementation environments are defining components of the AI interventions' overall potential effectiveness. Therefore, bringing AI systems from mathematical performance to clinical utility needs an adapted, stepwise implementation and evaluation pathway, addressing the complexity of this collaboration between two independent forms of intelligence, beyond measures of effectiveness alone 1 . Despite indications that some AI-based algorithms now match the accuracy of human experts within preclinical in silico studies 2 , there

Renal ischemia reperfusion injury (IRI), a common event after renal transplantation, causes acute kidney injury (AKI), increases the risk of delayed graft function (DGF), primes the donor kidney for rejection, and contributes to the long-term risk of graft loss. In the last decade, epidemiological studies have linked even mild episodes of AKI to chronic kidney disease (CKD) progression, and innate immunity seems to play a crucial role. The ischemic insult triggers an acute inflammatory reaction that is elicited by Pattern Recognition Receptors (PRRs), expressed on both infiltrating immune cells as well as tubular epithelial cells (TECs). Among the PRRs, Toll-like receptors (TLRs), their synergistic receptors, Nod-like receptors (NLRs), and the inflammasomes, play a pivotal role in shaping inflammation and TEC repair, in response to renal IRI. These receptors represent promising targets to modulate the extent of inflammation, but also function as gatekeepers of tissue repair, protecting against AKI-to-CKD progression. Despite the important considerations on timely use of therapeutics, in the context of IRI, treatment options are limited by a lack of understanding of the intra-and intercellular mechanisms associated with the activation of innate immune receptors and their impact on adaptive tubular repair. Accumulating evidence suggests that TEC-associated innate immunity shapes the tubular response to stress through the regulation of immunometabolism. Engagement of innate immune receptors provides TECs with the metabolic flexibility necessary for their plasticity during injury and repair. This could significantly affect pathogenic processes within TECs, such as cell death, mitochondrial damage, senescence, and pro-fibrotic cytokine secretion, well-known to exacerbate inflammation and fibrosis. This article provides an overview of the past 5 years of research on the role of innate immunity in experimental and human IRI, with a focus on the cascade of events activated by hypoxic damage in TECs: from programmed cell death (PCD) and Tammaro et al. Innate Immunity in Renal IRI mitochondrial dysfunction-mediated metabolic rewiring of TECs to maladaptive repair and progression to fibrosis. Finally, we will discuss the important crosstalk between metabolism and innate immunity observed in TECs and their therapeutic potential in both experimental and clinical research.

No abstract

Supplemental Digital Content is available in the text.

In renal transplantation, use of calcineurin inhibitors (CNIs) is associated with nephrotoxicity and immunosuppression with malignancies and infections. This trial aimed to minimize CNI exposure and total immunosuppression while maintaining efficacy. We performed a randomized controlled, open-label multicenter trial with early cyclosporine A (CsA) elimination. Patients started with basiliximab, prednisolone (P), mycophenolate sodium (MPS), and CsA. At 6 months, immunosuppression was tapered to P/CsA, P/MPS, or P/everolimus (EVL). Primary outcomes were renal fibrosis and inflammation. Secondary outcomes were estimated glomerular filtration rate (eGFR) and incidence of rejection at 24 months. The P/MPS arm was prematurely halted. The trial continued with P/CsA (N = 89) and P/EVL (N = 96). Interstitial fibrosis and inflammation were significantly decreased and the eGFR was significantly higher in the P/EVL arm. Cumulative rejection rates were 13% (P/EVL) and 19% (P/CsA), (p = 0.08). A post hoc analysis of HLA and donor-specific antibodies at 1 year after transplantation revealed no differences. An individualized immunosuppressive strategy of early CNI elimination to dual therapy with everolimus was associated with decreased allograft fibrosis, preserved allograft function, and good efficacy, but also with more serious adverse events and discontinuation. This can be a valuable alternative regimen in patients suffering from CNI toxicity.