Glioblastoma Multiforme (GBM) is a highly aggressive brain cancer characterized by rapid proliferation and extensive remodeling of the extracellular matrix (ECM), leading to progressive tissue stiffening. Although ECM stiffness is known to promote GBM progression, the molecular mechanisms linking mechanical cues to tumor growth remain insufficiently defined. In this study, transcriptomic comparison of GBM tumors and non-neoplastic brain tissue revealed coordinated upregulation of cell cycle regulators and matrisome-associated genes, with survivin (BIRC5) identified as a central node linking proliferative signaling and ECM remodeling networks. Analysis of GBM patient specimens further showed strong nuclear survivin expression in regions with elevated collagen deposition. To directly evaluate stiffness-dependent regulation of survivin, GBM cells were cultured on fibronectin-infused hydrogels with tunable stiffness. Stiff matrices increased survivin expression along with cyclin D1 and cyclin A, consistent with increased cell cycle progression. Pharmacologic inhibition or siRNA-mediated suppression of survivin reduced stiffness-induced proliferation and attenuated expression of matrisome components, including collagens and lysyl oxidase. These findings indicate that survivin functions as a mechanosensitive regulator that coordinates cell cycle progression with ECM production in stiff tumor microenvironments. Collectively, this study identifies survivin as a key mediator linking ECM stiffness to GBM growth and matrisome remodeling. Targeting survivin and its effectors may offer a mechanosensitive strategy to limit GBM growth.

Background:Donor-derived cell-free DNA is an established blood-based biomarker used to assess alloimmune activity after kidney transplantation. Methods:We performed a retrospective study of 257 kidney transplant recipients that had at least one dd-cfDNA measurements during 3-year period. The primary aim was to assess the association between dd-cfDNA levels with graft function and survival. Secondary exploratory aims included examining the relationships between dd-cfDNA strata and biopsy-proven rejection, donor-specific antibodies, C4d deposition, and longitudinal eGFR trajectories. Patients were stratified by their highest dd-cfDNA measurement into three groups: <0.50%, 0.50-0.99%, and ≥1.0%. Results:Patients categorized in the ≥1.0% dd-cfDNA group had increased rates of rejection, more severe histopathologic injury, and a higher prevalence of DSAs. With the ≥ 0.50-0.90% dd-cfDNA group having greater variability and decline in eGFR over the 3 year period. In exploratory multivariable modeling, higher dd-cfDNA strata were associated with a decline in graft function and survival. Conclusions:Elevated dd-cfDNA levels were associated with adverse alloimmune and functional outcomes, including rejection, DSA positivity, and reduced graft survival (p=0.0071). A logistic regression model identified eGFR decline in the >1.0% group to predict long-term graft failure and patient survival (p=0.04). These findings support the clinical value of dd-cfDNA as a biomarker of alloimmune risk in kidney transplant recipients.

Foundational models (FMs) based on advanced neural network architectures have demonstrated improved performance in pathology image analysis across various organs due to their increased generalizability. However, their clinical adoption requires explainability, as their black box nature limits transparency. Understanding the specific features these models learn for a given downstream task is crucial for explainability and integrating FMs into clinical workflows more effectively. We propose a computational pipeline that enhances explainability by correlating domain-specific handcrafted features (HFs), with hidden features i.e., feature embeddings (FEs) from FMs. We correlate and combine HFs from Detectron 2 DeepLabv3+ segmentation with FEs from Prov-Gigapath (PG) and UNI FMs for improved explainability and performance. In this work, HFs are extracted from segmented functional tissue units, including arteries, tubules, globally sclerotic glomeruli, and non-globally sclerotic glomeruli. FEs are extracted at the tile and slide levels for PG and at the tile level for UNI. We use the Pearson correlation coefficient to identify significant correspondences between these feature sets. To evaluate our proposed methodology, we use 56 diabetic nephropathy kidney biopsy whole slide images (WSIs) from Seoul National University Hospital. The task is to predict end-stage kidney disease (ESKD) two years post-biopsy using leave-one-out cross-validation on 56 WSIs, with 16 from ESKD patients and 40 from non-ESKD patients. We combine top correlated features from FEs of FMs with HFs and train logistic regression (LR) and k nearest neighbor (kNN) classifiers. LR model trained on combined feature set improved accuracy, balanced accuracy, Matthew's correlation coefficient, F1-score, precision, and recall to 0.8393, 0.7938, 0.5993, 0.8377, 0.8367, 0.8393 respectively, when compared to LR and kNN models trained on individual feature sets. PG excelled in specificity (1.000) and AUROC (0.8281), while UNI showed superior AUPRC (0.7813) performance. We also present feature explainability maps corresponding to each feature in FE.

Kidney transplantation is a critical intervention for managing end-stage renal disease, with post-transplant biopsies providing essential insights into graft function. This study aimed to develop and evaluate an automated system for classifying kidney biopsies taken immediately after transplantation to predict patient outcomes, specifically changes in estimated glomerular filtration rate (eGFR). We utilized a dataset of 56 Formalin-Fixed, Paraffin-Embedded (FFPE) whole slide images of time-zero biopsies, stained with Periodic Acid-Schiff (PAS), along with time series eGFR data recorded at 3 months, 6 months and 12 motnhs post-transplant. We employed ComPRePS (an in-house multicompartment segmentation tool) to analyze biopsy images, segmenting compartments such as cortical and medullary interstitium, non-globally and globally sclerotic glomeruli, tubules, and arteries/arterioles. Focus was placed on glomeruli, from which embeddings were extracted using a pretrained Vision Transformer (ViT) model (vit-base-patch16-224-in21k). These embeddings captured critical features for training an Artificial Neural Network (ANN) model to classify patients based on the percent change in eGFR from 3 months to 12 months, categorizing them into 'eGFR decline' and 'eGFR stable' groups. The proposed classification model achieved an accuracy of 0.73 in predicting eGFR changes. We validated these results by visualizing the two patient groups in a reduced dimensional domain using UMAP (Uniform Manifold Approximation and Projection), which revealed a clear distinction between the groups. This result underscores the potential of early biopsies for predicting long-term graft outcomes, enhancing patient management by providing early insights. We also trained a baseline ANN model using glomeruli hand-crafted features from ComPRePS, which achieved an accuracy of 0.55. Our proposed model achieved higher accuracy than baseline. Further research is needed to refine the model, expand the dataset, and validate findings across diverse populations to improve prediction accuracy and clinical applicability.

Spatial technologies examining the cell and tissue microenvironment at near single-cell resolution are revealing important molecular insights. However, few tools enable integrated, interactive analysis of spatial-omics with tissue morphology in the same functional tissue unit. Here, we present FUSION (Functional Unit State Identification in Whole Slide Images), a web-based platform for visualizing and analyzing spatial-omics data with high-resolution histology. FUSION provides workflows for assessing cell compositions, quantitative morphometrics, and comparative tissue analyses. We demonstrate applicability across spatial assays, including 10x Visium, Visium HD, 10x Xenium, Cell DIVE, and PhenoCycler, applied to healthy and diseased tissues from kidney, small intestine, lung, and skin in the Human BioMolecular Atlas Program. FUSION is cloud-based, open-source, and accessible at https://fusion.hubmapconsortium.org/ , hosting over 50 paired datasets and tutorials. In a series of use cases, we show its capacity to distinguish renal glomeruli injury states, quantify morphometric changes, and characterize fibrosis with immune infiltration.

Multi-omics data, such as 10X Genomics Visium (spatial transcriptomics), measure gene expressions, molecular pathway activities, and can predict cell types/states but are often expensive and inaccessible in clinical settings. Thus, despite the emergence of multi-omics technologies, histopathological assessments under brightfield microscopy remain the diagnostic gold standard. In this work, we examine machine learning-based pipelines for predicting cell types/states from brightfield histology images using state-of-the-art (SOTA) deep learning (DL) models, aiming to enhance diagnostics and prognostics in clinical medicine. Our proposed pipeline consists of two stages: (1) an Image-To-Text retrieval Network (ITTN) that leverages the CONtrastive learning from Captions for Histopathology (CONCH) model to assign histopathological text prompt from brightfield histology image, and (2) a Vision Language Model (VLM), which is built on the same CONCH model used in ITTN but incorporates a regression head to predict cell type/state proportions based on the paired image and text inputs. During training, we classify the image into one of four structural types (glomerulus, tubules, vessels, and interstitium) using the ITTN. These classification labels are then used to construct a new text prompt with a suitable histopathological description for each image in the test set. The new text prompt and raw image are used as paired inputs to the VLM to predict cell types/states. We also utilize SOTA models, such as CONCH (using only the vision encoder), ViT, and ResNet, which employ image-only inputs in separate regression pipelines. We experimented and tested our proposed pipelines on a set of 10X Visium formalin-fixed paraffin-embedded whole slides images of diabetic nephropathy samples collected at Indiana University. Our experiments yielded a mean squared error of 0.0027 for the proposed pipeline, showing improvements of 20.59%, 27.03%, and 32.50% over CONCH (image only), ViT, and ResNet, respectively. The proposed pipeline aims to bridge the gap between traditional histopathology and molecular diagnostics, enhancing disease diagnosis and prognosis.

HistoLens is an open-source graphical user interface developed using MATLAB AppDesigner for visual and quantitative analysis of histological datasets. HistoLens enables users to interrogate sets of digitally annotated whole slide images to efficiently characterize histological differences between disease and experimental groups. Users can dynamically visualize the distribution of 448 hand-engineered features quantifying color, texture, morphology, and distribution across microanatomic sub-compartments. Additionally, users can map differentially detected image features within the images by highlighting affected regions. We demonstrate the utility of HistoLens to identify hand-engineered features that correlate with pathognomonic renal glomerular characteristics distinguishing diabetic nephropathy and amyloid nephropathy from the histologically unremarkable glomeruli in minimal change disease. Additionally, we examine the use of HistoLens for glomerular feature discovery in the Tg26 mouse model of HIV-associated nephropathy. We identify numerous quantitative glomerular features distinguishing Tg26 transgenic mice from wild-type mice, corresponding to a progressive renal disease phenotype. Thus, we demonstrate an off-the-shelf and ready-to-use toolkit for quantitative renal pathology applications.

Spatial -OMICS technologies facilitate the interrogation of molecular profiles in the context of the underlying histopathology and tissue microenvironment. Paired analysis of histopathology and molecular data can provide pathologists with otherwise unobtainable insights into biological mechanisms. To connect the disparate molecular and histopathologic features into a single workspace, we developed FUSION (Functional Unit State IdentificatiON in WSIs [Whole Slide Images]), a web-based tool that provides users with a broad array of visualization and analytical tools including deep learning-based algorithms for in-depth interrogation of spatial -OMICS datasets and their associated high-resolution histology images. FUSION enables end-to-end analysis of functional tissue units (FTUs), automatically aggregating underlying molecular data to provide a histopathology-based medium for analyzing healthy and altered cell states and driving new discoveries using "pathomic" features. We demonstrate FUSION using 10x Visium spatial transcriptomics (ST) data from both formalin-fixed paraffin embedded (FFPE) and frozen prepared datasets consisting of healthy and diseased tissue. Through several use-cases, we demonstrate how users can identify spatial linkages between quantitative pathomics, qualitative image characteristics, and spatial --omics

Chronic kidney disease (CKD) is a global health concern, with its progression often characterized by pathological changes in renal tubules. Accurate segmentation and quantitative analysis of renal tubules are critical for understanding disease progression and monitoring treatment efficacy in both model studies and routine diagnostics. Here, we present the development and testing of a novel software tool designed for Automated Renal Tubular Segmentation and Analysis (ARTSA). The ARTSA software employs advanced deep learning algorithms and computational image analysis to automate the segmentation and analysis of renal tubules and medulla in histological kidney tissue sections. This innovative approach eliminates the need for labor-intensive manual segmentation and minimizes human bias, thereby enhancing the precision and efficiency of renal tubular analysis. To gauge ARTSA’s performance, we carefully tested it on a balanced dataset of kidney tissue images, covering both healthy and diseased states. The software demonstrated exceptional segmentation performance for both renal tubules and nuclei. Furthermore, the ARTSA software enables comprehensive quantitative analysis of segmented tubules, including measures of tubular curvature, brush border loss, and luminal expansion, which are essential pathologies to quantify. In summary, the ARTSA software represents a significant contribution to the field of digital renal pathology by offering a reliable, automated solution for renal tubular segmentation and analysis.

The Banff Digital Pathology Working Group (DPWG) was established with the goal to establish a digital pathology repository; develop, validate, and share models for image analysis; and foster collaborations using regular videoconferencing. During the calls, a variety of artificial intelligence (AI)-based support systems for transplantation pathology were presented. Potential collaborations in a competition/trial on AI applied to kidney transplant specimens, including the DIAGGRAFT challenge (staining of biopsies at multiple institutions, pathologists’ visual assessment, and development and validation of new and pre-existing Banff scoring algorithms), were also discussed. To determine the next steps, a survey was conducted, primarily focusing on the feasibility of establishing a digital pathology repository and identifying potential hosts. Sixteen of the 35 respondents (46%) had access to a server hosting a digital pathology repository, with 2 respondents that could serve as a potential host at no cost to the DPWG. The 16 digital pathology repositories collected specimens from various organs, with the largest constituent being kidney ( n = 12,870 specimens). A DPWG pilot digital pathology repository was established, and there are plans for a competition/trial with the DIAGGRAFT project. Utilizing existing resources and previously established models, the Banff DPWG is establishing new resources for the Banff community.

[This corrects the article DOI: 10.1017/cts.2023.172.].

Background:The heterogeneous phenotype of diabetic nephropathy (DN) from type 2 diabetes complicates appropriate treatment approaches and outcome prediction. Kidney histology helps diagnose DN and predict its outcomes, and an artificial intelligence (AI)-based approach will maximize clinical utility of histopathological evaluation. Herein, we addressed whether AI-based integration of urine proteomics and image features improves DN classification and its outcome prediction, altogether augmenting and advancing pathology practice.Methods:We studied whole slide images (WSIs) of periodic acid-Schiff-stained kidney biopsies from 56 DN patients with associated urinary proteomics data. We identified urinary proteins differentially expressed in patients who developed end-stage kidney disease (ESKD) within two years of biopsy. Extending our previously published human-AI-loop pipeline, six renal sub-compartments were computationally segmented from each WSI. Hand-engineered image features for glomeruli and tubules, and urinary protein measurements, were used as inputs to deep-learning frameworks to predict ESKD outcome. Differential expression was correlated with digital image features using the Spearman rank sum coefficient.Results:A total of 45 urinary proteins were differentially detected in progressors, which was most predictive of ESKD (AUC=0.95), while tubular and glomerular features were less predictive (AUC=0.71 and AUC=0.63, respectively). Accordingly, a correlation map between canonical cell-type proteins, such as epidermal growth factor and secreted phosphoprotein 1, and AI-based image features was obtained, which supports previous pathobiological results.Conclusions:Computational method-based integration of urinary and image biomarkers may improve the pathophysiological understanding of DN progression as well as carry clinical implications in histopathological evaluation.

OBJECTIVES/GOALS: Computational pathology is an emerging discipline that resides at the intersection of engineering, computer science, and pathology. There is a growing need to develop innovative pedagogical approaches to train future computational pathologists who have diverse educational backgrounds. METHODS/STUDY POPULATION: Our work proposes an iterative approach toward teaching master’s and Ph.D. students from various backgrounds, such as electrical engineering, biomedical engineering, and cell biology the basics of cell-type identification. This approach is grounded in the active learning framework to allow for observation, reflection, and independent application. The learners are trained by a team of an electrical engineer and pathologist and provided with eight images containing a glomerulus. They must then classify nuclei in each of the glomeruli as either a podocyte (blue), endothelial cell (green), or mesangial cell (red). RESULTS/ANTICIPATED RESULTS: A simple web application was built to calculate agreement, measured using Cohen’s kappa, between annotators for both individual glomeruli and across all eight images. Automating the process of providing feedback from an expert renal pathologist to the learner allows for learners to quickly determine where they can improve. After initial training, agreement scores for cells scored by both the learner and the expert were high (0.75), however, when including cells not scored by both the agreement was relatively low (0.45). This indicates that learners needed more instruction on identifying unique cells within each image. This low-stakes approach encourages exploratory and generative learning. DISCUSSION/SIGNIFICANCE: Computation medical sciences require interdisciplinary training methods. We report on a robust approach for team-based mentoring and skill development. Future implementations will include undergraduate learners and provide opportunities for graduate students to engage in near-peer mentoring.

The incorporation of automated computational tools has a great amount of potential to positively influence the field of pathology. However, pathologists and regulatory agencies are reluctant to trust the output of complex models such as Convolutional Neural Networks (CNNs) due to their usual implementation as black-box tools. Increasing the interpretability of quantitative analyses is a critical line of research in order to increase the adoption of modern Machine Learning (ML) pipelines in clinical environments. Towards that goal, we present HistoLens, a Graphical User Interface (GUI) designed to facilitate quantitative assessments of datasets of annotated histological compartments. Additionally, we introduce the use of hand-engineered feature visualizations to highlight regions within each structure that contribute to particular feature values. These feature visualizations can then be paired with feature hierarchy determinations in order to view which regions within an image are significant to a particular sub-group within the dataset. As a use case, we analyzed a dataset of old and young mouse kidney sections with glomeruli annotated. We highlight some of the functional components within HistoLens that allow non-computational experts to efficiently navigate a new dataset as well as allowing for easier transition to downstream computational analyses.

Histological image data and molecular profiles provide context into renal condition. Often, a biopsy is drawn to diagnose or monitor a suspected kidney problem. However, molecular profiles can go beyond a pathologist’s ability to see and diagnose. Using AI, we computationally incorporated urinary proteomic profiles with microstructural morphology from renal biopsy to investigate new and existing molecular links to image phenotypes. We studied whole slide images of periodic acid-Schiff stained renal biopsies from 56 DN patients matched with 2,038 proteins measured from each patient’s urine. Using Seurat, we identified differentially expressed proteins in patients that developed end-stage renal disease within 2 years of biopsy. Glomeruli, globally sclerotic glomeruli, and tubules were segmented from WSI using our previously published HAIL pipeline. For each glomerulus, 315 handcrafted digital image features were measured, and for tubules, 207 features. We trained fully connected networks to predict urinary protein measurements that were differentially expressed between patients who did/ did not progress to ESRD within 2 years of biopsy. The input to this network was either glomerular or tubular histomorphological features in biopsy. Trained network weights were used as a proxy to rank which morphological features correlated most highly with specific urinary proteins. We identified significant image feature-protein pairs by ranking network weights by magnitude. We also looked at which features on average were most significant in predicting proteins. For both glomeruli and tubules, RGB color values and variance in PAS+ areas (specifically basement membrane for tubules) were, on average, more predictive of molecular profiles than other features. There is a strong connection between molecular profile and image phenotype, which can be elucidated through computational methods. These discovered links can provide insight to disease pathways, and discover new factors contributing to incidence and progression.

Background Toward development of diagnostics for cryptogenic stroke, we hypothesize that histomic features of stroke blood clots retrieved by mechanical thrombectomy could be used to delineate stroke etiology. Methods Clots were retrieved from patients undergoing thrombectomy, and etiology was determined by the trial of TOAST (Trial of Org 10172 in Acute Stroke Treatment) score. After sectioning and hematoxylin and eosin staining, clot components (red blood cells [RBCs], fibrin–platelet aggregates [FPs], and white blood cells [WBCs]) were segmented on whole slide images. Histomic features were engineered to capture structural distribution of RBC/FP regions, including radiomics, radial composition, and RBC/FP object features. To locally characterize WBCs, textural features derived from nuclear and extranuclear regions were computed from each WBC to define classes, which we summarized into class frequency distributions. Univariate and multivariate statistics were used to identify significant differences in engineered features between large artery atherosclerosis (LAA) and cardioembolic cases. The top 3 significant RBC/FP and WBC features were used to train a complement Naïve Bayes model, which was then used to predict the etiology of cryptogenic cases. Results In our data (n=53), 31 clots were cardioembolic, 8 were LAA, 4 were of strokes of other determined etiology, and 10 were cryptogenic. We identified 17 significant RBC/FP features and 3 significant WBC class frequency distributions that were different between cardioembolic and LAA. A complement Naïve Bayes model accurately classified cardioembolic versus LAA with a validation area under the receiver operating characteristic curve of 0.87±0.03, a performance substantially higher to using clot component percent composition (area under the receiver operating characteristic curve=0.69±0.16) that is the current state‐of‐the‐art. Further, cryptogenic cases were reliably classified as cardioembolic or LAA in cross‐validation analysis. Conclusion We present a first‐of‐its‐kind histomics pipeline to robustly quantify the complex structure and WBC heterogeneity in acute ischemic stroke clots and classify cryptogenic cases. We hope this work begins to pave the way for histopathology biomarkers for stroke etiology diagnosis.

Introduction: Determining stroke etiology is paramount to clinical management and prevention of recurrent strokes. Although stroke patients undergo extensive post-treatment work-ups, 30-40% of cases remain cryptogenic. Hypothesis: Engineered histomic features from digital pathology images of stroke blood clots collected during mechanical thrombectomy (MT) can be used to delineate stroke etiology. Methods: For clots retrieved from patients undergoing MT, etiology was determined by the trial of Trial of Org 10172 in Acute Stroke Treatment (TOAST) score. After sectioning and H&E staining, clot components (red blood cells-RBCs, fibrin-platelet regions-FP, and white blood cells-WBCs) were segmented from whole slide images. Histomic features were engineered to capture the structural distribution of RBC/FP regions throughout the clot. To measure clot WBC diversity, WBC instances were clustered into WBC “classes” based on texture, and summarized as a class frequency distribution (CFD). The three most significant RBC/FP and WBC features between large artery atherosclerosis (LAA) and cardioembolic (CE) cases were used to train a complement Naïve Bayes (CNB) model, which was then implemented to predict the etiology of cryptogenic cases. Results: In our data (n=53), 31 clots were CE, 8 were LAA, 4 were from strokes of other determined etiology, and 10 were cryptogenic. 17 significant RBC/FP features and 3 significant WBC CFDs were different between CE and LAA. The trained CNB model accurately classified CE vs. LAA with a validation AUC of 0.87±0.03, exhibiting superior performance to a model trained using common clot percent cellular composition metrics (AUC=0.69±0.16). Furthermore, we showed the potential to classify cryptogenic cases as CE or LAA. Conclusions: This first-of-its-kind, biologically-informed clot histomic pipeline captured significant information that may augment clinical and laboratory features used in stroke etiology classification.