A Deep Learning Instance Segmentation Approach for Global Glomerulosclerosis Assessment in Donor Kidney Biopsies

The histological assessment of glomeruli is fundamental for determining if a kidney is suitable for transplantation. The Karpinski score is essential to evaluate the need for a single or dual kidney transplant and includes the ratio between the number of sclerotic glomeruli and the overall number of glomeruli in a kidney section. The manual evaluation of kidney biopsies performed by pathologists is time-consuming and error-prone, so an automatic framework to delineate all the glomeruli present in a kidney section can be very useful. Our experiments have been conducted on a dataset provided by the Department of Emergency and Organ Transplantations (DETO) of Bari University Hospital. This dataset is composed of 26 kidney biopsies coming from 19 donors. The rise of Convolutional Neural Networks (CNNs) has led to a realm of methods which are widely applied in Medical Imaging. Deep learning techniques are also very promising for the segmentation of glomeruli, with a variety of existing approaches. Many methods only focus on semantic segmentation—which consists in segmentation of individual pixels—or ignore the problem of discriminating between non-sclerotic and sclerotic glomeruli, so these approaches are not optimal or inadequate for transplantation assessment. In this work, we employed an end-to-end fully automatic approach based on Mask R-CNN for instance segmentation and classification of glomeruli. We also compared the results obtained with a baseline based on Faster R-CNN, which only allows detection at bounding boxes level. With respect to the existing literature, we improved the Mask R-CNN approach in sliding window contexts, by employing a variant of the Non-Maximum Suppression (NMS) algorithm, which we called Non-Maximum-Area Suppression (NMAS). The obtained results are very promising, leading to improvements over existing literature. The baseline Faster R-CNN-based approach obtained an F-Measure of 0.904 and 0.667 for non-sclerotic and sclerotic glomeruli, respectively. The Mask R-CNN approach has a significant improvement over the baseline, obtaining an F-Measure of 0.925 and 0.777 for non-sclerotic and sclerotic glomeruli, respectively. The proposed method is very promising for the instance segmentation and classification of glomeruli, and allows to make a robust evaluation of global glomerulosclerosis. We also compared Karpinski score obtained with our algorithm to that obtained with pathologists’ annotations to show the soundness of the proposed workflow from a clinical point of view.

3326 Radiofrequency Renal Denervation Prevents Further Progression of Hypertension and Decreases Renal Medullary Fibrosis in One-year-old Spontaneously Hypertensive Rats (SHR)

OBJECTIVES/SPECIFIC AIMS: We have reported that radiofrequency renal denervation (RF-RDN) in SHR at 20-weeks of age, decreased blood pressure (BP) and fibrosis in kidney cortex and medulla when rats were sacrificed at 6 months. However, whether RF-RDN can have similar benefits in older rats remains unknown. This study examined whether performing RF-RDN in older rats also has a beneficial effect on BP and renal fibrosis. METHODS/STUDY POPULATION: Baseline systolic and diastolic BP (SBP/DPB) was measured (telemetry) in nine-month-old SHR and Wistar Kyoto rats (WKY). Groups of rats then received bilateral RF-RDN or Sham-RDN (SHR-RDN, n=9; SHR-Sham, n=10; WKY-RDN, n=5; WKY-Sham, n=8). Rats were then sacrificed at 12-months of age. Kidneys were harvested, sectioned, and assessed for fibrosis by Masson’s trichrome stain. A pathologist, who was blinded to treatment groups, evaluated each kidney section for fibrosis. RESULTS/ANTICIPATED RESULTS: Compared to SHR with Sham-RDN, RF-RDN prevented a further increase in systolic and diastolic BP from baseline (9-month) in SHR as they aged to 12-months (SHR-Sham mmHg: 9-month 193±4/127±4; 12-month 207±3/142±5; SHR-RDN mmHg: 9-month 197±3/132±2; 12-month 197±4/132±3). RF-RDN did not alter SBP or DBP in aged WKY. One-year-old SHR with prior Sham-RDN showed extensive renal fibrosis in kidney cortex and medulla. In contrast, RF-RDN significantly decreased renal fibrosis in the medulla, but not cortex. There was no fibrosis in kidneys of age matched WKY. DISCUSSION/SIGNIFICANCE OF IMPACT: These findings suggest that RF-RDN may be a potential therapy for halting progression of hypertension and decreasing medullary fibrosis in the aged population.

Effect of Time-Lapse Administration of Panadol (Paracetamol) on Spleen and Kidney Functions of Adult Albino Mice

Panadol is a remarkable pain and fever reducing non opioid drug. It is known to be completely safe and tolerant medicine throughout the globe among people of all age groups. The goal of this study was to elaborate the histopathological effects of reduced interrupted regime of panadol on spleen and kidney functions of mice (Mus musculus). The animals were categorized into four groups, the control group (C), and rest of the three were labeled as 1, 2 and 3, made on the basis of time interval of pandol administration via gavage (n=7). Matched volume of panadol (15000 µg/0.1ml) and normal saline was given to mice. Tissue samples were collected after sacrifice of the mice and processed for assessment. The findings of the current study reflected the histopathological damage of kidney and spleen caused by panadol in reduced interval of time. The kidney section illustrated clear distortion in glomeruli integrity, marked increase in interstitial spaces, damaged epithelia, and degeneration in tubules in all the groups. The spleen histology exhibited the degradation of white pulp, depopulation, activation of follicles, cellular disruption thereby overall disorganized stature. The raised values of serum creatinine and blood urea examinations also revealed the deleterious effects of panadol overconsumption. It is inferred from the above mentioned outcomes that though panadol is considered to be a safe drug even then its intake prior to four hours can account for adverse effects on kidney and spleen.

Molecular imaging of small molecule drugs in animal tissues using laser desorption postionization mass spectrometry

Localization and quantification of ACF in a kidney section by laser desorption postionization mass spectrometry imaging.

Abstract 010: CD44 and CD44+ Cells are Dispensable for the Recruitment of Renin Expressing Cells

In response to a homeostatic stress the number of cells that make renin increases dramatically along the renal arteriolar tree resembling the embryonic pattern. We have shown that this “recruitment” occurs by re-expression of renin in smooth muscle cells that differentiated from embryonic renin cells. A recent study proposed that during recruitment, renal CD44+ mesenchymal stem-like cells can differentiate into juxtaglomerular (JG)-like renin-producing cells. To test such hypothesis, we assessed the distribution and role of CD44+ cells in renin cell recruitment. Mice with homozygous (KO) and heterozygous (het) deletion of CD44 (knockin for LacZ) were treated with low-sodium diet (0.05%) plus captopril (0.5 g/l) for 10 days (n: 9 treated, 7 controls). Body and kidney weights and BP were not different between KO and het mice. BUN and creatinine were significantly increased in both KO and Het treated mice. The number of renin expressing cells in the kidney and circulating renin increased similarly in treated mice (ELISA, untreated: het 131,503 ± 19,319 pg/mL vs KO 84,714 ± 29,065 pg/mL p=0.2517; treated: het 367,850 ± 38,189 pg/mL vs KO 495,120 ± 80,311 pg/mL p=0.2311). Interestingly, immunostaining for CD44 was negative in kidneys of untreated and treated wild type mice. We occasionally observed in CD44-LacZ het or KO mice isolated LacZ positive cells inside the glomeruli (1 or less per sagittal kidney section) and none in the JG area. On the other hand, immunostaining for CD44 on kidney sections of Ren1cKO mice revealed positive cells within perivascular infiltrates. To confirm these results we performed qRT-PCR for CD44 on kidney samples from CD44 het and KO treated, untreated, control, and Ren1cKO mice. CD44 mRNA expression confirmed the histological findings. In summary: 1) CD44 is dispensable for renin expression and recruitment, and 2) CD44+ cells do not contribute to the pool of renin expressing cells in the kidney during basal conditions or in response to a homeostatic stress as previously suggested. However, they do participate in the inflammatory process observed surrounding the vessels in mice with deletion of the renin gene, suggesting that they derived from the circulation and not from the kidney.

Revealing a Pre-neoplastic Renal Tubular Lesion by p-S6 Protein Immunohistochemistry after Rat Exposure to Aristolochic Acid

Aristolochic acid (AA) has, in the last decade, become widely promoted as the cause of the Balkan endemic nephropathy and associated renal or urothelial tumours, although without substantial focal evidence of the quantitative dietary exposure via bread in specific households in hyperendemic villages. Occasional ethnobotanical use of Aristolochia clematitis might be a source of AA, and Pliocene lignite contamination of well-water is also a putative health risk factor. The aim of this study was two-fold: to verify if extracts of A. clematitis and Pliocene, or AA by itself, could induce the development of renal or urothelial tumours, and to test the utility of the ribosomal protein p-S6 to identify preneoplastic transformation. Rats were given extracts of A. clematitis in drinking water or AA I, by gavage. After seven months, renal morphology was studied using conventional haematoxylin and eosin and immunohistochemistry for ribosomal p-S6 protein. Plant extracts (cumulative AA approximately 1.8 g/kg b.w.) were tolerated and caused no gross pathology or renal histopathological change, with only faint diffuse p-S6 protein (except in the papilla) as in controls. Cumulative AA I (150 mg/kg b.w. given over 3 days) was also tolerated for seven months by all recipients, without gross pathology or kidney tumours. However, p-S6 protein over-expression was consistent particularly within the renal papilla. In one case given AA I, intense p-S6 protein staining of a proximal tubule fragment crucially matched the pre-neoplastic histology in an adjacent kidney section. We briefly discuss these findings, which compound uncertainty concerning the cause of the renal or upper urinary tract tumours of the Balkan endemic nephropathy.

Case 11

This is a case from the Kidney section of Genitourinary Imaging Cases. The specific content of each case and its diagnosis are necessarily hidden from this abstract. Each case contains case history, followed immediately by the diagnosis, which is supported with findings, differential diagnosis, teaching points, management, and further reading suggestions.

Case 10

This is a case from the Kidney section of Genitourinary Imaging Cases. The specific content of each case and its diagnosis are necessarily hidden from this abstract. Each case contains case history, followed immediately by the diagnosis, which is supported with findings, differential diagnosis, teaching points, management, and further reading suggestions.

Case 2

This is a case from the Kidney section of Genitourinary Imaging Cases. The specific content of each case and its diagnosis are necessarily hidden from this abstract. Each case contains case history, followed immediately by the diagnosis, which is supported with findings, differential diagnosis, teaching points, management, and further reading suggestions.

Case 29

This is a case from the Kidney section of Genitourinary Imaging Cases. The specific content of each case and its diagnosis are necessarily hidden from this abstract. Each case contains case history, followed immediately by the diagnosis, which is supported with findings, differential diagnosis, teaching points, management, and further reading suggestions.