Smartphone-based machine learning model for real-time assessment of medical kidney biopsy.

Background: Kidney biopsy is the gold-standard for diagnosing medical renal diseases, but the accuracy of the diagnosis greatly depends on the quality of the biopsy specimen, particularly the amount of renal cortex obtained. Inadequate biopsies, characterized by insufficient cortex or predominant medulla, can lead to inconclusive or incorrect diagnoses, and repeat biopsy. Unfortunately, there has been a concerning increase in the rate of inadequate kidney biopsies, and not all medical centers have access to trained professionals who can assess biopsy adequacy in real time. In response to this challenge, we aimed to develop a machine learning model capable of assessing the percentage cortex of each biopsy pass using smartphone images of the kidney biopsy tissue at the time of biopsy. Methods: 747 kidney biopsy cores and corresponding smartphone macro images were collected from five unused deceased donor kidneys. Each core was imaged, formalin-fixed, sectioned, and stained with Periodic acid-Schiff (PAS) to determine cortex percentage. The fresh unfixed core images were captured using the macro camera on an iPhone 13 Pro. Two experienced renal pathologists independently reviewed the PAS-stained sections to determine the cortex percentage. For the purpose of this study, the biopsies with less than 30% cortex were labeled as inadequate, while those with 30% or more cortex were classified as adequate. The dataset was divided into training (n=643), validation (n=30), and test (n=74) sets. Preprocessing steps involved converting High-Efficiency Image Container iPhone format images to JPEG, normalization, and renal tissue segmentation using a U-Net deep learning model. Subsequently, a classification deep learning model was trained on the renal tissue region of interest and corresponding class label. Results: The deep learning model achieved an accuracy of 85% on the training data. On the independent test dataset, the model exhibited an accuracy of 81%. For inadequate samples in the test dataset, the model showed a sensitivity of 71%, suggesting its capability to identify cases with inadequate cortical representation. The area under the receiver-operating curve (AUC-ROC) on the test dataset was 0.80. Conclusion: We successfully developed and tested a machine learning model for classifying smartphone images of kidney biopsies as either adequate or inadequate, based on the amount of cortex determined by expert renal pathologists. The model's promising results suggest its potential as a smartphone application to assist real-time assessment of kidney biopsy tissue, particularly in settings with limited access to trained personnel. Further refinements and validations are warranted to optimize the model's performance.

A web-based tool for real-time adequacy assessment of kidney biopsies.

The escalating incidence of kidney biopsies providing insufficient tissue for diagnosis poses a dual challenge, straining the healthcare system and jeopardizing patients who may require rebiopsy or face the prospect of an inaccurate diagnosis due to an unsampled disease. Here, we introduce a web-based tool that can provide real-time, quantitative assessment of kidney biopsy adequacy directly from photographs taken with a smartphone camera. The software tool was developed using a deep learning-driven automated segmentation technique, trained on a dataset comprising nephropathologist-confirmed annotations of the kidney cortex on digital biopsy images. Our framework demonstrated favorable performance in segmenting the cortex via 5-fold cross-validation (Dice coefficient: 0.788±0.130) (n=100). Offering a bedside tool for kidney biopsy adequacy assessment has the potential to provide real-time guidance to the physicians performing medical kidney biopsies, reducing the necessity for re-biopsies. Our tool can be accessed through our web-based platform: http://www.biopsyadequacy.org.

Clinicopathologic Features of Antibrush Border Antibody Disease.

Introduction: Antibrush border antibody disease (ABBA) is an autoimmune tubulointerstitial kidney disease that primarily affects older individuals and results in progressive kidney failure. It is rare with only 20 reported cases. Here, we describe a case series to further define the clinicopathologic spectrum and natural history, and to inform management. Methods: We identified 67 patients with ABBA who underwent kidney biopsy, including 65 native and 2 transplants. Demographics, clinical findings, and laboratory data were obtained. Histopathologic data included light microscopy, immunofluorescence, electron microscopy and immunostaining for LRP2, CUBN, and AMN. Follow-up data, including treatment(s), laboratory values, and outcomes, were available from 51 patients. Results: Patients with ABBA were predominantly male with a median age of 72 years. Median serum creatinine was 2.7 mg/dl, proteinuria was 2.8 g/day, and hematuria was present in two-thirds of the patients. Tubular injury with LRP2-positive tubular basement membrane (TBM) deposits were seen in 94.2% of patients. Thirty-eight patients (56.7%) had a second kidney disease, commonly glomerular diseases with high-grade proteinuria. These diseases included podocytopathies, membranous nephropathy (MN), IgA nephropathy, diabetic glomerulopathy, lupus nephritis (LN), crescentic glomerulonephritis (GN), tubulointerstitial nephritis, and involvement by lymphoma. The majority of patients were treated with immunosuppression. Of those patients with follow-up, 29.4% achieved remission, 70.6% had no response, and 52.8% required dialysis or were deceased. Untreated patients were at the highest risk. Conclusion: ABBA is a rare autoimmune kidney disease that often occurs with other kidney diseases. Although the overall prognosis of ABBA is poor, there is potential benefit from immunosuppression.

Development and Validation of a Multiclass Model Defining Molecular Archetypes of Kidney Transplant Rejection: A Large Cohort Study of the Banff Human Organ Transplant Gene Expression Panel.

Gene expression profiling from formalin-fixed paraffin-embedded (FFPE) renal allograft biopsies is a promising approach for feasibly providing a molecular diagnosis of rejection. However, large-scale studies evaluating the performance of models using NanoString platform data to define molecular archetypes of rejection are lacking. We tested a diverse retrospective cohort of over 1400 FFPE biopsy specimens, rescored according to Banff 2019 criteria and representing 10 of 11 United Network of Organ Sharing regions, using the Banff Human Organ Transplant panel from NanoString and developed a multiclass model from the gene expression data to assign relative probabilities of 4 molecular archetypes: No Rejection, Antibody-Mediated Rejection, T Cell-Mediated Rejection, and Mixed Rejection. Using Least Absolute Shrinkage and Selection Operator regularized regression with 10-fold cross-validation fitted to 1050 biopsies in the discovery cohort and technically validated on an additional 345 biopsies, our model achieved overall accuracy of 85% in the discovery cohort and 80% in the validation cohort, with ≥75% positive predictive value for each class, except for the Mixed Rejection class in the validation cohort (positive predictive value, 53%). This study represents the technical validation of the first model built from a large and diverse sample of diagnostic FFPE biopsy specimens to define and classify molecular archetypes of histologically defined diagnoses as derived from Banff Human Organ Transplant panel gene expression profiling data.

Kidney Pathology Education for Nephrology Fellows: Past, Present, and Future.

Kidney pathology education is a critical component in training of nephrology fellows, as well as for continuing medical education for practicing nephrologists. Kidney pathology images are included on nephrology fellow board exams, and clinicopathologic correlation of kidney biopsy findings is critical in everyday clinical practice. Nephropathology training is a requirement by the American College of Graduate Medical Education within nephrology fellowship curricula. However, greater than one-third of fellowship program directors believe that nephropathology training for their fellows is not sufficient. During the Coronavirus Disease-19 pandemic, the use of digital learning has become commonplace with virtual conferences (local, national, and international) and online meetings becoming the norm for education. Nephrology has become a leader in free open-access online medical education, both prior to and, to even a greater extent, during the pandemic. Here, we review available resources to nephrology fellows and other learners to supplement nephropathology training, which includes medical blogs, journal clubs, interactive quizzes and games, online conferences, podcasts, and mentorship opportunities. These resources are archived and provide durable content to learners of all stages of training, even beyond the pandemic.

Boosting Quantum Battery-Based IoT Gadgets via RF-Enabled Energy Harvesting.

The search for a highly portable and efficient supply of energy to run small-scale wireless gadgets has captivated the human race for the past few years. As a part of this quest, the idea of realizing a Quantum battery (QB) seems promising. Like any other practically tractable system, the design of QBs also involve several critical challenges. The main problem in this context is to ensure a lossless environment pertaining to the closed-system design of the QB, which is extremely difficult to realize in practice. Herein, we model and optimize various aspects of a Radio-Frequency (RF) Energy Harvesting (EH)-assisted, QB-enabled Internet-of-Things (IoT) system. Several RF-EH modules (in the form of micro- or nano-meter-sized integrated circuits (ICs)) are placed in parallel at the IoT receiver device, and the overall correspondingly harvested energy helps the involved Quantum sources achieve the so-called quasi-stable state. Concretely, the Quantum sources absorb the energy of photons that are emitted by a photon-emitting device controlled by a micro-controller, which also manages the overall harvested energy from the RF-EH ICs. To investigate the considered framework, we first minimize the total transmit power under the constraints on overall harvested energy and the number of RF-EH ICs at the QB-enabled wireless IoT device. Next, we optimize the number of RF-EH ICs, subject to the constraints on total transmit power and overall harvested energy. Correspondingly, we obtain suitable analytical solutions to the above-mentioned problems, respectively, and also cross-validate them using a non-linear program solver. The effectiveness of the proposed technique is reported in the form of numerical results, which are both theoretical and simulations based, by taking a range of operating system parameters into account.

Hybrid Active-and-Passive Relaying Model for 6G-IoT Greencom Networks with SWIPT.

In order to support a massive number of resource-constrained Internet-of-Things (IoT) devices and machine-type devices, it is crucial to design a future beyond 5G/6G wireless networks in an energy-efficient manner while incorporating suitable network coverage expansion methodologies. To this end, this paper proposes a novel two-hop hybrid active-and-passive relaying scheme to facilitate simultaneous wireless information and power transfer (SWIPT) considering both time-switching (TS) and power-splitting (PS) receiver architectures, while dynamically modelling the involved dual-hop time-period (TP) metric. An optimization problem is formulated to jointly optimize the throughput, harvested energy, and transmit power of a SWIPT-enabled system with the proposed hybrid scheme. In this regard, we provide two distinct ways to obtain suitable solutions based on the Lagrange dual technique and Dinkelbach method assisted convex programming, respectively, where both the approaches yield an appreciable solution within polynomial computational time. The experimental results are obtained by directly solving the primal problem using a non-linear optimizer. Our numerical results in terms of weighted utility function show the superior performance of the proposed hybrid scheme over passive repeater-only and active relay-only schemes, while also depicting their individual performance benefits over the corresponding benchmark SWIPT systems with the fixed-TP.

A multi-center retrospective cohort study defines the spectrum of kidney pathology in Coronavirus 2019 Disease (COVID-19).

Kidney failure is common in patients with Coronavirus Disease-19 (COVID-19), resulting in increased morbidity and mortality. In an international collaboration, 284 kidney biopsies were evaluated to improve understanding of kidney disease in COVID-19. Diagnoses were compared to five years of 63,575 native biopsies prior to the pandemic and 13,955 allograft biopsies to identify diseases that have increased in patients with COVID-19. Genotyping for APOL1 G1 and G2 alleles was performed in 107 African American and Hispanic patients. Immunohistochemistry for SARS-CoV-2 was utilized to assess direct viral infection in 273 cases along with clinical information at the time of biopsy. The leading indication for native biopsy was acute kidney injury (45.4%), followed by proteinuria with or without concurrent acute kidney injury (42.6%). There were more African American patients (44.6%) than patients of other ethnicities. The most common diagnosis in native biopsies was collapsing glomerulopathy (25.8%), which was associated with high-risk APOL1 genotypes in 91.7% of cases. Compared to the five-year biopsy database, the frequency of myoglobin cast nephropathy and proliferative glomerulonephritis with monoclonal IgG deposits was also increased in patients with COVID-19 (3.3% and 1.7%, respectively), while there was a reduced frequency of chronic conditions (including diabetes mellitus, IgA nephropathy, and arterionephrosclerosis) as the primary diagnosis. In transplants, the leading indication was acute kidney injury (86.4%), for which rejection was the predominant diagnosis (61.4%). Direct SARS-CoV-2 viral infection was not identified. Thus, our multi-center large case series identified kidney diseases that disproportionately affect patients with COVID-19 and demonstrated a high frequency of APOL1 high-risk genotypes within this group, with no evidence of direct viral infection within the kidney.

NELL1 is a target antigen in malignancy-associated membranous nephropathy.

Patients with membranous nephropathy have an increased risk of malignancy compared to the general population, but the target antigen for malignancy-associated membranous nephropathy is unknown. To explore this, we utilized mass spectrometry for antigen discovery in malignancy-associated membranous nephropathy examining immune complexes eluted from frozen kidney biopsy tissue using protein G bead immunoglobulin capture. Antigen discovery was performed comparing cases of membranous nephropathy of unknown and known type. Mass spectrophotometric analysis revealed that nerve epidermal growth factor-like 1 (NELL1) immune complexes were uniquely present within the biopsy tissue in membranous nephropathy. Additional NELL1-positive cases were subsequently identified by immunofluorescence. In a consecutive series, 3.8% of PLA2R- and THSD7A-negative cases were NELL1-positive. These NELL1-positive cases had segmental to incomplete IgG capillary loop staining (93.4%) and dominant or co-dominant IgG1-subclass staining (95.5%). The mean age of patients with NELL1-positive membranous nephropathy was 66.8 years, with a slight male predominance (58.2%) and 33% had concurrent malignancy. Compared with PLA2R- and THSD7A-positive cases of membranous nephropathy, there was a greater proportion of cases with malignancies in the NELL1-associated group. Thus, NELL1-associated membranous nephropathy has a unique histopathology characterized by incomplete capillary loop staining, IgG1-predominance, and is more often associated with malignancy than other known types of membranous nephropathy.

Neural cell adhesion molecule 1 is a novel autoantigen in membranous lupus nephritis.

Membranous lupus nephritis is a frequent cause of nephrotic syndrome in patients with systemic lupus erythematosus. It has been shown in phospholipase A2 receptor positive membranous nephropathy that known antibodies can be detected within sera, determination of the target autoantigen can have diagnostic significance, inform prognosis, and enable non-invasive monitoring of disease activity. Here we utilized mass spectrometry for antigen discovery in laser captured microdissected glomeruli from formalin-fixed paraffin embedded tissue and tissue protein G immunoprecipitation studies to interrogate immune complexes from frozen kidney biopsy tissue. We identified neural cell adhesion molecule 1 (NCAM1) to be a target antigen in some cases of membranous lupus nephritis and within rare cases of primary membranous nephropathy. The prevalence of NCAM1 association was 6.6% of cases of membranous lupus nephritis and in 2.0% of primary membranous nephropathy cases. NCAM1 was found to colocalize with IgG within glomerular immune deposits by confocal microscopy. Additionally, serum from patients with NCAM1-associated membranous nephropathy showed reactivity to NCAM1 recombinant protein on Western blotting and by indirect immunofluorescence assay, demonstrating the presence of circulating antibodies. Thus, we propose that NCAM1 is a target autoantigen in a subset of patients with membranous lupus nephritis. Future studies are needed to determine whether anti-NCAM1 antibody levels correlate with disease activity or response to therapy.

A Uromodulin Mutation Drives Autoimmunity and Kidney Mononuclear Phagocyte Endoplasmic Reticulum Stress.

We identified a family with a UMOD gene mutation (C106F) resulting in glomerular inflammation and complement deposition. To determine if the observed phenotype is due to immune system activation by mutant uromodulin, a mouse strain with a homologous cysteine to phenylalanine mutation (C105F) in the UMOD gene was generated using CRISPR-Cas9 gene editing and the effect of this mutation on mononuclear phagocytic cells was examined. Mutant mice developed high levels of intracellular and secreted aggregated uromodulin, resulting in anti-uromodulin antibodies and circulating uromodulin containing immune complexes with glomerular deposition and kidney fibrosis with aging. F4/80+ and CD11c+ kidney cells phagocytize uromodulin. Differential gene expression analysis by RNA sequencing of F4/80+ phagocytic cells revealed activation of the activating transcription factor 5 (ATF5)-mediated stress response pathway in mutant mice. Phagocytosis of mutant uromodulin by cultured dendritic cells resulted in activation of the endoplasmic reticulum stress response pathway and markers of cell inactivation, an effect not seen with wild-type protein. Mutant mice demonstrate a twofold increase in T-regulatory cells, consistent with induction of immune tolerance, resulting in decreased inflammatory response and improved tissue repair following ischemia-reperfusion injury. The C105F mutation results in autoantibodies against aggregated misfolded protein with immune complex formation and kidney fibrosis. Aggregated uromodulin may induce dendritic cell tolerance following phagocytosis through an unfolded protein/endoplasmic reticulum stress response pathway, resulting in decreased inflammation following tissue injury.

Detection of SARS-CoV-2 in formalin-fixed paraffin-embedded tissue sections using commercially available reagents.

Coronavirus Disease-19 (COVID-19), caused by the coronavirus SARS-CoV-2, was initially recognized in Wuhan, China and subsequently spread to all continents. The disease primarily affects the lower respiratory system, but may involve other organs and systems. Histopathologic evaluation of tissue from affected patients is crucial for diagnostic purposes, but also for advancing our understanding of the disease. For that reason, we developed immunohistochemical (IHC) and in situ hybridization (ISH) assays for detection of the. virus. A total of eight autopsy lungs, one placenta, and ten kidney biopsies from COVID-19 patients were stained with a panel of commercially available antibodies for IHC and commercially available RNA probes for ISH. Similarly, autopsy lungs, placentas and renal biopsies from non-COVID-19 patients were stained with the same antibodies and probes. All eight lungs and the placenta from COVID-19 patients stained positive by IHC and ISH, while the kidney biopsies stained negative by both methodologies. As expected, all specimens from non-COVID-19 patients were IHC and ISH negative. These two assays represent a sensitive and specific method for detecting the virus in tissue samples. We provide the protocols and the list of commercially available antibodies and probes for these assays, so they can be readily implemented in pathology laboratories and medical examiner offices for diagnostic and research purposes.

Serum amyloid P deposition is a sensitive and specific feature of membranous-like glomerulopathy with masked IgG kappa deposits.

Membranous-like glomerulopathy with masked IgG kappa deposits (MGMID) is a recently described pattern of glomerulonephritis with a unique histopathology. The pattern is characterized by subepithelial and/or mesangial immune deposits that are "masked", to immunoglobulin staining by routine immunofluorescence but strongly stain for IgG and kappa light chain after protease digestion. Patients with this pattern of glomerulonephritis are most commonly young females presenting with proteinuria and a vague history of autoimmune disease such as low titer antinuclear antibodies. Here we compared the mass spectrometry profile of laser capture microdissected glomeruli from nine MGMID renal biopsies with eight biopsies showing other patterns of membranous glomerulopathy. The protein most significantly increased in MGMID was serum amyloid P. Immunostaining showed serum amyloid P colocalized with IgG in the glomeruli of MGMID but not with PLA2R-associated membranous glomerulopathy. Serum amyloid P was positive in the glomeruli of all 32 MGMID biopsies but negative in biopsies of other types of membranous glomerulopathies such as those associated with PLA2R and THSD7A. There were four biopsies with glomerular serum amyloid P staining among the 173 biopsies that did not fulfill criteria for MGMID or amyloidosis. All four of these biopsies with positive serum amyloid P staining had a membranous pattern of glomerulopathy with IgG kappa deposits that only differed from MGMID by the lack of "masking". Thus, positive staining within glomerular deposits for serum amyloid P identifies a unique form of glomerulonephritis likely sharing a common pathophysiologic mechanism of disease.

Id1 expression in kidney endothelial cells protects against diabetes-induced microvascular injury.

The inhibitor of differentiation (Id) transcription regulators, which are induced in response to oxidative stress, promote cell proliferation and inhibit senescence. Inhibitor of differentiation 1 (Id1) expression is limited to endothelial cells (EC) in the normal mouse kidney and is required for a normal response to injury. Endothelial dysfunction leads to the development of diabetic nephropathy, and so, we hypothesized that endothelial Id1 may help protect against hyperglycemia-induced microvascular injury and nephropathy. Here, we tested this hypothesis by using streptozotocin to induce diabetes in Id1 knockout (KO) mice and WT B6;129 littermates and examining the mice at 3 months. Expression of Id1 was observed to be increased 15-fold in WT kidney EC, and Id1 KO mice exhibited increased mesangial and myofibroblast proliferation, matrix deposition, and albuminuria compared with WT mice. Electron microscopy demonstrated peritubular capillary EC injury and lumen narrowing, and fluorescence microangiography showed a 45% reduction in capillary perfusion area with no reduction in CD31-stained areas in Id1 KO mice. Microarray analysis of EC isolated from WT and KO control and diabetic mice demonstrated activation of senescence pathways in KO cells. Kidneys from KO diabetic mice showed increased histological expression of senescence markers. In addition, premature senescence in cultured KO EC was also seen in response to oxidative stress. In conclusion, endothelial Id1 upregulation with hyperglycemia protects against microvascular injury and senescence and subsequent nephropathy.

Tissue staining for THSD7A in glomeruli correlates with serum antibodies in primary membranous nephropathy: a clinicopathological study.

Thrombospondin type-1 domain-containing 7A (THSD7A) is the most recently recognized target antigen in patients with membranous nephropathy. We stained membranous nephropathy biopsies processed in our laboratory for phospholipase A2 receptor and THSD7A over an 18-month period and selected all THSD7A-positive cases for study. Serum samples from most patients were tested by an indirect immunofluorescence assay for the presence of THSD7A antibodies (Euroimmun). A total of 31 patients were diagnosed with THSD7A-associated membranous nephropathy for a prevalence of 2.4% among patients with membranous nephropathy. The patients were most often male (male-to-female ratio of 1.6) with a mean age of 62 years and a mean proteinuria of 9.6 g per day (range 1.1-15.9). Two of the 31 patients had a history of cancer and none were diagnosed with malignancy on follow-up. Serum samples were available at the time of biopsy from 24 patients and all tested positive for antibodies against THSD7A. Conversely, all 20 serum samples from patients with membranous nephropathy who had negative staining for THSD7A were negative for serum reactivity to THSD7A. We conclude that THSD7A tissue staining of kidney biopsies with membranous nephropathy is a sensitive and specific method for the diagnosis of THSD7A-associated membranous nephropathy and it correlates strongly with the serum antibody testing. We also present the clinicopathologic details of the largest cohort to date of THSD7A-associated membranous nephropathy from a single institution.

An integrated pathology and ultrasonography-based simulation for training in performing kidney biopsy
.

BACKGROUND: Medical practice trends and limitations in trainees' duty hours have diminished the interest and exposure of nephrology fellows to percutaneous kidney biopsy (PKB). We hypothesized that an integrated nephrology-pathology-led simulation may be an effective educational tool. MATERIALS AND METHODS: A 4-hour PKB simulation workshop (KBSW), led by two ultrasonography (US)-trained nephrologists and two nephropathologists, consisted of 6 stations: 1) diagnostic kidney US with live patients, 2) kidney pathology with plasticine models of embedded torso cross-sections, 3) US-based PKB with mannequin (Blue Phantom™), 4) kidney pathology with dissected cadavers, 5) US-based PKB in lightly-embalmed cadavers, and 6) tissue retrieval adequacy examination by microscope. A 10-question survey assessing knowledge acquisition and procedural confidence gain was administered pre- and post-KBSW. RESULTS: 21 participants attended the KBSW and completed the surveys. The overall percentage of correct answers to knowledge questions increased from 55 to 83% (p = 0.016). The number of "extremely confident" answers increased from 0 - 5% to 19 - 28% in all 4 questions (p = 0.02 - 0.04), and the number of "not at all confident" answers significantly decreased from 14 - 62% to 0 - 5% in 3 out of 4 questions (p = 0.0001 - 0.03). Impact of the imparted training on subsequent practice pattern was not assessed. CONCLUSION: A novel KBSW is an effective educational tool to acquire proficiency in PKB performance and could help regain interest among trainees in performing PKBs.
.