Histological Slides Research Articles

A multi-modal artificial intelligence model trained on H&E tissue images and clinical patient data for prediction of BRAF mutations in colorectal cancer.

e15161 Background: Analyzing genetic mutations is essential for personalized therapy decisions in oncology and beyond. In colorectal cancer (CRC), mutations in BRAF ( BRAFmut) render patients eligible for targeted treatment with BRAF inhibitors and immunotherapy. BRAFmut is routinely analyzed via genomic sequencing, which, however, is costly and does not provide diagnostic results immediately. Reliable detection of BRAFmut status from histological images using artificial intelligence (AI) could provide a more accessible and faster diagnostic alternative. Previous approaches to image-based BRAFmut analysis, however, failed to reach a critical level of accuracy on independent test cohorts, with reported area under the receiver operating characteristic curve (AUROC) values typically below 0.8. Methods: We hypothesized that combining tissue image analysis with clinical patient data could improve diagnostic accuracy for BRAFmut. Following our hypothesis, we developed a multi-modal AI model on n = 426 CRC cases from the TCGA database using BRAFmut data, H&E-stained tissue images, and information on patient parameters that are readily available in clinical practice and that are known to be associated with BRAFmut frequency. The model was based on a vision transformer architecture incorporating pathology-specific information using a pretrained embedding model for encoding tissue image features. Model performance was validated on an independent hold-out TCGA cohort of n = 107 samples and an additional external cohort of n = 104 CRC samples from the CPTAC database. Results: Our multi-modal AI model reached an AUROC of 0.85 ± 0.02 on the TCGA hold-out test set. Notably, on the second, external CPTAC dataset the predictive accuracy level was similar (AUROC of 0.83 ± 0.02) indicating the robustness of our approach to generalize to different datasets. Additional analyses showed that the significant performance boost over previous approaches was gained by the multi-modal analysis setup, as predicting BRAFmut based on either tissue, or clinical data alone led to lower accuracy values. Explainability analysis of AI results confirmed that relevant biological features on the histological slides, mainly tumor areas, were considered by the model for decision making. Conclusions: The combination of H&E images with clinically readily available patient parameters enabled us to develop a model predicting BRAFmut in CRC that surpasses previous approaches in diagnostic accuracy on independent evaluation cohorts. This indicates that combining histological tissue information with clinical parameters can increase the ability of AI models to predict genetic variants from tissue and more generally highlights the potential of multi-modal deep learning approaches to enhance predictive AI models in digital pathology.

Enabling accessibility to the CellScape spatial biology workflow for researchers using standard histology protocols and pre-slided specimens.

e15185 Background: Solid tumor specimen collection during clinical trials often follows a standard histology protocol, ending in the generation of FFPE sections mounted onto 1 mm histology slides. Though a common sample format for researchers developing tissue-based biomarkers, slides present a data quality challenge for multiplexed spatial biology workflows. Here we present data describing an innovative flow cell design fully compatible with pre-slided tissue specimens. We collected spatial biology data using this slide-compatible flow cell and compared it to the gold standard direct mounting on coverglass. Several rounds of marker staining and imaging demonstrate equivalency in method outcomes including image resolution, signal to noise, and re-interrogation capacity. Methods: FFPE biopsy tissue specimens (tumor punch biopsies and a tonsil section) were mounted on standard histology slides. These specimens were incorporated into flow cells using a new glass coverslip and porting device, CellScape™ Slide, that pairs with a standard histology slide to form a microfluidic chamber. After mounting, each sample was probed for consensus CD markers (CD45, CD3, CD4,CD8,CD20, etc.) through several rounds of iterative immunofluorescence staining using CellScape™ Precise Spatial Multiplexing. The resulting multi-layered whole slide image dataset was then interrogated for spatial resolution and signal-to-noise using digital image analysis. Results: Spatial resolution and signal-to-noise ratio (SNR) were assessed through comparison to data collected from similar samples using the existing CellScape microfluidic chip. The lowest intensity signal collected with at least a 30:1 SNR (acceptable for image analysis) was equivalent or lower in the CellScape Slide preparation (equivalent sensitivity). The highest intensity signal collected was equivalent between the two flow cell designs. Comparing stain quality (specificity, sensitivity) among markers stained through multiple rounds of iterative staining, imaging, and signal remove via photoinactivation showed that the CellScape Slide flow cell is on par with or better than the existing CellScape chip. Further, the usable clear window of the CellScape Slide flow chamber, measuring 16 mm x 44 mm, was found to be greater than twice that of the existing chip. Literature review determined this to also be the largest imaging window available among commercially available spatial biology platforms. Conclusions: Assessment of the CellScape Slide flow cell, innovative in providing access to the CellScape spatial biology workflow for pre-mounted FFPE tissue, determined that the optical performance of the new flow cell is at least on par with the existing flow cell. Further, the usable window is over twice the area, and the new flow cell is fully capable of supporting re-interrogation of the sample.

Effect of parental ploidy on reproductive system development in sturgeon hybrids in the second year of life

The paper presents the data on germ cell development in various sturgeon hybrids with various ploidy levels. The study was performed at experimental industrial sites of the Branch for Freshwater Fisheries of the Russian Federal Research Institute of Fisheries and Oceanography (VNIIPRKH). Histology samples were taken in the winter, summer and autumn. Histology slides were made following the standard procedure. The figures show all stages of germ cell development in males and females of various hybrids, depending on rearing conditions (thermal constant). Two-year-old Russian sturgeon × Kaluga fish hybrids with similar ploidy levels reached gonad maturity stage I at a thermal constant of 8,571 degree-days (547 days). Female Russian sturgeon × beluga hybrids aged 631 days (10,191 degree-days) showed accelerated germ cell development characteristic of stage I to early stage II. Male Russian sturgeon × beluga and Siberian sturgeon × beluga hybrids had sexually indifferent germ cells. These species are expected to be sterile. Male Russian sturgeon × Kaluga fish hybrids aged 374 days had sexually indifferent germ cells with occasional primary spermatogonia; cell division was observed on Day 547. The study found that distant hybrids show better growth than hybrids with similar ploidy levels, where more intensive germ cell development is observed. Further research will elucidate the effect of gonad development during reproductive system formation on the growth of sturgeon hybrids with various ploidy levels.

Abstract B004: Deep learning unveils molecular footprints in histology: predicting molecular subtypes from bladder cancer histology slides

Abstract Introduction: Transurethral resection followed by intravesical bacillus Calmette-Guerin (BCG) is the standard of care for high-risk non-muscle-invasive bladder cancer (HR-NMIBC) patients. However, ≈20% of HR-NMIBC patients progress to muscle-invasive or metastatic disease, which is associated with a high-risk of death. Early identification of patients who are unlikely to benefit from BCG treatment could benefit from an early radical cystectomy or bladder-sparing alternatives. Previously, we showed that molecular subtyping or HR-NMIBC identified BCG response subtypes (BRS 1/2/3). BRS3 tumors were associated with a poor response to BCG and a high-risk of progression. However, molecular subtyping relies on resource-intensive and costly sequencing methods such as next-generation sequencing, and microarray analysis. We employed deep learning (DL) to predict BRS3 vs BRS1/2 molecular subtypes from digitized histology slides. Methods: Haematoxylin and eosin (H&E)-stained slides were digitized, and quality control was performed. The areas where the RNA was extracted were annotated pixel-wise by a trained pathology assistant, using QuPath. Image tiles of 512 × 512 pixels at 10X, 20X, and 40X magnifications were extracted with a 25% overlap. For classification, BRS1 and BRS2 images were grouped, while BRS3 images were categorized separately. After applying Macenko normalization, tiles were split based on individual patients into a 75/25 training/validation ratio via stratified 4-fold cross-validation. Four DL models (DenseNet, Inception, ShuffleNet, ResNet) were trained to classify tiles into BRS3 and BRS1/2, with tile labels based on the patient’s subtype. Patient-level classification was performed using majority voting from tile predictions. Model efficacy was gauged using the area under the ROC curve (AUC). Results: Out of 245 H&E slides, 45 were discarded after quality control. From 200 patients – with a 70:30 distribution between BRS1&2 and BRS3 The DenseNet model at 10x magnification was the best-performing model, achieving the highest AUC compared to all other models and magnifications. On the image tile level, the best DenseNet achieved an average AUC of 0.65 with a standard deviation of 0.03 across the four folds of the validation set. At the patient-level prediction, the best DenseNet reached an average AUC of 0.80 and a standard deviation of 0.05 on the validation set. Conclusion: Our study demonstrates that deep learning classifies BSR3 from BSR1 and 2 subtypes from H&E slides. This method holds potential to identify BRS3 HR-NMIBC patients who may benefit from alternative treatments than BCG. With further refinement, this deep learning strategy can become a key tool in histology, aiding in swift, affordable, and efficient subtype identification. Hence, this method has the potential to bridge the gap between molecular techniques and clinical practice. Future research should prioritize evaluating the model on an external test cohort and consider predicting additional molecular markers to fine-tune prediction outcomes. Citation Format: Farbod Khoraminia, Flouros C de Jong, Farhan Akram, Geert Litjens, Maarten D.J. Jansen, Alberto Nakuama Gonzalez, Danique Lichtenburg, Andrew Stubbs, Nadieh Khalili, Tahlita C.M. Zuiverloon. Deep learning unveils molecular footprints in histology: predicting molecular subtypes from bladder cancer histology slides [abstract]. In: Proceedings of the AACR Special Conference on Bladder Cancer: Transforming the Field; 2024 May 17-20; Charlotte, NC. Philadelphia (PA): AACR; Clin Cancer Res 2024;30(10_Suppl):Abstract nr B004.

Extraskeletal Ewing Sarcoma of the Gastrointestinal and Hepatobiliary Tract: Deceptive Immunophenotype Commonly Leads to Misdiagnosis.

Ewing sarcoma (ES) is an uncommon mesenchymal neoplasm that typically develops as a bone mass, although up to 30% arise in extraskeletal sites. ES of the gastrointestinal (GI) and hepatobiliary tract is rare and may be misdiagnosed as other, more common neoplasms that occur in these sites. However, the correct classification of extraskeletal ES is important for timely clinical management and prognostication. We reviewed our experience of ES in the GI and hepatobiliary tract in order to further highlight the clinicopathologic features of these neoplasms and document the potential for misdiagnosis in this setting. The archives and consultation files of 6 academic institutions were retrospectively queried for cases of ES occurring in the GI and hepatobiliary tract. The histologic slides and ancillary studies were reviewed and clinical data were retrieved for each case through the electronic medical records, when available. Twenty-three patients with ES in the GI and/or hepatobiliary tract were identified from 2000 to 2022. Of these, 11 were women and 12 were men with a median age of 38 years (range, 2 to 64). Tumor locations included the pancreas (n=5), liver (n=2), stomach (n=3), colorectum (n=3), and small intestine (n=5), as well as tumors involving multiple organs, pelvis and retroperitoneum (n=5). Tumor size varied between 2cm and 18cm. Twenty were primary and 3 were metastases. Of the 23 cases, only 17% were initially diagnosed as ES. The most common misdiagnoses involved various forms of neuroendocrine neoplasia due to expression of synaptophysin and other neuroendocrine markers (22%). A wide variety of diagnoses including GI stromal tumor was considered due to aberrant CD117 expression (4%). The diagnosis of ES was ultimately confirmed by detection of the EWSR1 rearrangement in 22 cases. The remaining case was diagnosed using traditional immunohistochemistry. Follow-up information was available in 20 cases, with follow-up time varying between 2 and 256 months. Six patients with follow-up died of disease between 6 and 60 months following initial presentation. Our data indicate ES in the GI and hepatobiliary tract is commonly misdiagnosed leading to a delay in therapy. In light of the attendant therapeutic and prognostic implications, ES should be considered in the differential diagnosis of any GI or hepatobiliary tumor with epithelioid and/or small round cell morphology.

Digital histological staining of tissue slide images from optical coherence microscopy.

The convergence of staining-free optical imaging and digital staining technologies has become a central focus in digital pathology, presenting significant advantages in streamlining specimen preparation and expediting the rapid acquisition of histopathological information. Despite the inherent merits of optical coherence microscopy (OCM) as a staining-free technique, its widespread application in observing histopathological slides has been constrained. This study introduces a novel approach by combining wide-field OCM with digital staining technology for the imaging of histopathological slides. Through the optimization of the histology slide production process satisfying the ground growth for digital staining as well as pronounced contrast for OCM imaging, successful imaging of various mouse tissues was achieved. Comparative analyses with conventional staining-based bright field images were executed to evaluate the proposed methodology's efficacy. Moreover, the study investigates the generalization of digital staining color appearance to ensure consistent histopathology, considering tissue-specific and thickness-dependent variations.

Histologic Hitchhikers: A Review of Common Exogenous Artifacts Encountered During Mohs Micrographic Surgery.

Exogenous artifacts can interfere with accurate histologic tissue evaluation on frozen sections during Mohs micrographic surgery (MMS). Mohs surgeons should be aware of these anomalies to avoid potential misdiagnoses. To review exogenous artifacts encountered in frozen tissue pathology during MMS. A literature search was conducted in PubMed to identify studies reporting on exogenous artifacts encountered during MMS and a list of previously described exogenous artifacts was compiled. A retrospective examination of frozen histology slides from recent Mohs cases at the authors' institution was performed to obtain illustrative examples of these artifacts, supplemented by formalin-fixed paraffin-embedded samples when frozen examples could not be found. Exogenous artifacts represent foreign bodies that have been externally introduced into the skin or artifacts resulting from other external factors. If frozen section evaluation is inaccurate, overdiagnosis can occur during MMS, resulting in unnecessary layers, larger margins, and more complex reconstructions. These exogenous tissue changes can mimic inflammatory processes and melanocytic or keratinocyte malignancies on histology. Exogenous artifacts are common findings during margin assessment in Mohs micrographic surgery. The resulting histological findings can be confusing but correlating them with the clinical and surgical history often reassures surgeons. Recognizing these artifacts facilitates accurate diagnosis and promotes optimal patient care.

Cartilage Collagen Neoepitope C2C Expression in the Articular Cartilage and Its Relation to Joint Tissue Damage in Patients with Knee Osteoarthritis.

Pathological cleavage of type II collagen (Col2) and generation of Col2 neoepitopes can serve as useful molecular markers of the progression of osteoarthritis (OA). One of such potential biomarkers is type II collagen neoepitope C2C. The aim of this study was to correlate the degree of articular cartilage damage in OA patients with C2C expression in histological samples of tissues removed during total knee replacement. Cartilage samples were obtained from 27 patients ranging in age from 55 to 66 years. In each patient, medial and lateral tibia plateau samples were analyzed according to the OARSI histopathology grading system. The C2C expression was evaluated on histological slides by semi-quantitative analysis using ImageJ Fiji 2.14.0 software. Spearman's rank correlation analysis revealed a positive weak correlation (rho = 0.289, p = 0.0356) between the histological grade of tissue damage and the percentage of C2C staining. In addition, a highly significant positive correlation (rho = 0.388, p = 0.0041) was discovered between the osteoarthritis score (combining the histological grade of damage with the OA macroscopic stage) and the percentage of C2C staining in the samples. The C2C expression was detected in all the regions of the articular cartilage (i.e., the superficial zone, mid zone, deep zone and tidemark area, and the zone of calcified cartilage). Our findings imply that local expression of C2C correlates with the articular cartilage damage in OA-affected knees. This confirms that C2C can be used as a prospective marker for assessing pathological changes in the OA course and OA clinical trials.

Cyto R-CNN and CytoNuke Dataset: Towards reliable whole-cell segmentation in bright-field histological images

Background and objective:Cell segmentation in bright-field histological slides is a crucial topic in medical image analysis. Having access to accurate segmentation allows researchers to examine the relationship between cellular morphology and clinical observations. Unfortunately, most segmentation methods known today are limited to nuclei and cannot segment the cytoplasm. Methods:We present a new network architecture Cyto R-CNN that is able to accurately segment whole cells (with both the nucleus and the cytoplasm) in bright-field images. We also present a new dataset CytoNuke, consisting of multiple thousand manual annotations of head and neck squamous cell carcinoma cells. Utilizing this dataset, we compared the performance of Cyto R-CNN to other popular cell segmentation algorithms, including QuPath’s built-in algorithm, StarDist, Cellpose and a multi-scale Attention Deeplabv3+. To evaluate segmentation performance, we calculated AP50, AP75 and measured 17 morphological and staining-related features for all detected cells. We compared these measurements to the gold standard of manual segmentation using the Kolmogorov–Smirnov test. Results:Cyto R-CNN achieved an AP50 of 58.65% and an AP75 of 11.56% in whole-cell segmentation, outperforming all other methods (QuPath 19.46/0.91%; StarDist 45.33/2.32%; Cellpose 31.85/5.61%, Deeplabv3+ 3.97/1.01%). Cell features derived from Cyto R-CNN showed the best agreement to the gold standard (D¯=0.15) outperforming QuPath (D¯=0.22), StarDist (D¯=0.25), Cellpose (D¯=0.23) and Deeplabv3+ (D¯=0.33). Conclusion:Our newly proposed Cyto R-CNN architecture outperforms current algorithms in whole-cell segmentation while providing more reliable cell measurements than any other model. This could improve digital pathology workflows, potentially leading to improved diagnosis. Moreover, our published dataset can be used to develop further models in the future.

WhARIO: whole-slide-image-based survival analysis for patients treated with immunotherapy.

Immune checkpoint inhibitors (ICIs) are now one of the standards of care for patients with lung cancer and have greatly improved both progression-free and overall survival, although of the patients respond to the treatment, and some face acute adverse events. Although a few predictive biomarkers have integrated the clinical workflow, they require additional modalities on top of whole-slide images and lack efficiency or robustness. In this work, we propose a biomarker of immunotherapy outcome derived solely from the analysis of histology slides. We develop a three-step framework, combining contrastive learning and nonparametric clustering to distinguish tissue patterns within the slides, before exploiting the adjacencies of previously defined regions to derive features and train a proportional hazards model for survival analysis. We test our approach on an in-house dataset of 193 patients from 5 medical centers and compare it with the gold standard tumor proportion score (TPS) biomarker. On a fivefold cross-validation (CV) of the entire dataset, the whole-slide image-based survival analysis for patients treated with immunotherapy (WhARIO) features are able to separate a low- and a high-risk group of patients with a hazard ratio (HR) of 2.29 ( to 3.56), whereas the TPS 1% reference threshold only reaches a HR of 1.81 ( to 2.69). Combining the two yields a higher HR of 2.60 ( to 3.94). Additional experiments on the same dataset, where one out of five centers is excluded from the CV and used as a test set, confirm these trends. Our uniquely designed WhARIO features are an efficient predictor of survival for lung cancer patients who received ICI treatment. We achieve similar performance to the current gold standard biomarker, without the need to access other imaging modalities, and show that both can be used together to reach even better results.

An Update on the Use of Artificial Intelligence in Digital Pathology for Oral Epithelial Dysplasia Research.

Oral epithelial dysplasia (OED) is a precancerous histopathological finding which is considered the most important prognostic indicator for determining the risk of malignant transformation into oral squamous cell carcinoma (OSCC). The gold standard for diagnosis and grading of OED is through histopathological examination, which is subject to inter- and intra-observer variability, impacting accurate diagnosis and prognosis. The aim of this review article is to examine the current advances in digital pathology for artificial intelligence (AI) applications used for OED diagnosis. We included studies that used AI for diagnosis, grading, or prognosis of OED on histopathology images or intraoral clinical images. Studies utilizing imaging modalities other than routine light microscopy (e.g., scanning electron microscopy), or immunohistochemistry-stained histology slides, or immunofluorescence were excluded from the study. Studies not focusing on oral dysplasia grading and diagnosis, e.g., to discriminate OSCC from normal epithelial tissue were also excluded. A total of 24studies were included in this review. Nineteen studies utilized deep learning (DL) convolutional neural networks for histopathological OED analysis, and 4 used machine learning (ML) models. Studies were summarized by AI method, main study outcomes, predictive value for malignant transformation, strengths, and limitations. ML/DL studies for OED grading and prediction of malignant transformation are emerging as promising adjunctive tools in the field of digital pathology. These adjunctive objective tools can ultimately aid the pathologist in more accurate diagnosis and prognosis prediction. However, further supportive studies that focus on generalization, explainable decisions, and prognosis prediction are needed.

Prediction of Mismatch Repair Status in Endometrial Cancer from Histological Slide Images Using Various Deep Learning-Based Algorithms.

The application of deep learning algorithms to predict the molecular profiles of various cancers from digital images of hematoxylin and eosin (H&E)-stained slides has been reported in recent years, mainly for gastric and colon cancers. In this study, we investigated the potential use of H&E-stained endometrial cancer slide images to predict the associated mismatch repair (MMR) status. H&E-stained slide images were collected from 127 cases of the primary lesion of endometrial cancer. After digitization using a Nanozoomer virtual slide scanner (Hamamatsu Photonics), we segmented the scanned images into 5397 tiles of 512 × 512 pixels. The MMR proteins (PMS2, MSH6) were immunohistochemically stained, classified into MMR proficient/deficient, and annotated for each case and tile. We trained several neural networks, including convolutional and attention-based networks, using tiles annotated with the MMR status. Among the tested networks, ResNet50 exhibited the highest area under the receiver operating characteristic curve (AUROC) of 0.91 for predicting the MMR status. The constructed prediction algorithm may be applicable to other molecular profiles and useful for pre-screening before implementing other, more costly genetic profiling tests.

Structured illumination microscopy for cancer identification in diagnostic breast biopsies.

Breast cancer is the second most common cancer diagnosed in women in the US with almost 280,000 new cases anticipated in 2023. Currently, on-site pathology for location guidance is not available during the collection of breast biopsies or during surgical intervention procedures. This shortcoming contributes to repeat biopsy and re-excision procedures, increasing the cost and patient discomfort during the cancer management process. Both procedures could benefit from on-site feedback, but current clinical on-site evaluation techniques are not commonly used on breast tissue because they are destructive and inaccurate. Ex-vivo microscopy is an emerging field aimed at creating histology-analogous images from non- or minimally-processed tissues, and is a promising tool for addressing this pain point in clinical cancer management. We investigated the ability structured illumination microscopy (SIM) to generate images from freshly-obtained breast tissues for structure identification and cancer identification at a speed compatible with potential on-site clinical implementation. We imaged 47 biopsies from patients undergoing a guided breast biopsy procedure using a customized SIM system and a dual-color fluorescent hematoxylin & eosin (H&E) analog. These biopsies had an average size of 0.92 cm2 (minimum 0.1, maximum 4.2) and had an average imaging time of 7:29 (minimum 0:22, maximum 37:44). After imaging, breast biopsies were submitted for standard histopathological processing and review. A board-certified pathologist returned a binary diagnostic accuracy of 96% when compared to diagnoses from gold-standard histology slides, and key tissue features including stroma, vessels, ducts, and lobules were identified from the resulting images.

Galectin-3 expression by thyroid neoplasms in a center in southwest Nigeria

Background: Galectin-3 (GAL-3), which is known to be expressed in certain normal tissues and is implicated in tumor progression and metastasis, serves as a valuable marker for distinguishing malignant thyroid lesions from benign thyroid lesions. This study aimed to assess the clinical and histopathological features of thyroid neoplasms at OAUTHC Ile Ife utilizing GAL-3 immunohistochemistry and to compare these findings with those of studies conducted in other regions of the world. Materials and Methods: Histology slides from 56 patients with thyroid neoplasms spanning a 20-year period were subjected to immunohistochemical staining with monoclonal antibodies against GAL-3. Evaluation was based on the 2022 World Health Organization Classification of Thyroid Neoplasms, categorizing marker expression as follows: 1+ (10–25% stained cells), 2+ (26–50% stained cells), 3+ (51–75% stained cells), and 4+ (>75% stained cells). Negative staining indicated no staining, while localized staining (focal) represented <50% of cells stained (1+ or 2+), and diffuse staining indicated >50% of cells stained (3+ or 4+). Results: A total of 56 patients with neoplastic thyroid disease were studied, including 15 males and 41 females. Among the 16 benign neoplastic lesions, four (25.0%) exhibited focal GAL-3 expression, while 12 (75.0%) showed no expression. Among the forty malignant neoplastic lesions, five (12.5%) tested negative for the GAL-3 immunomarker, while 35 (87.5%) demonstrated positive expression to varying degrees. Among the positive malignant lesions, two (5.0%) displayed focal GAL-3 expression, while 33 (82.5%) exhibited diffuse staining of tumor cells. The diffuse and strong expression of GAL-3 was significantly greater in malignant thyroid neoplasms than in benign neoplasms (p<0.05). The sensitivity and specificity of GAL-3 were 88% and 75%, respectively. Conclusion: Our findings align with the previous research indicating that diffuse and strong immunohistochemical expression of GAL-3 is indicative of thyroid gland malignancy. GAL-3 serves as a valuable tool in diagnosing thyroid neoplasms, particularly in cases where histomorphologic characteristics are inconclusive or unclear.

Preliminary study of proton magnetic resonance spectroscopy to assess bone marrow adiposity in the third metacarpus or metatarsus in Thoroughbred racehorses.

Magnetic resonance spectroscopy (MRS) has been used to investigate metabolic changes within human bone. It may be possible to use MRS to investigate bone metabolism and fracture risk in the distal third metacarpal/tarsal bone (MC/MTIII) in racehorses. To determine the feasibility of using MRS as a quantitative imaging technique in equine bone by using the 1H spectra for the MC/MTIII to calculate fat content (FC). Observational cross-sectional study. Limbs from Thoroughbred racehorses were collected from horses that died or were subjected to euthanasia on racecourses. Each limb underwent magnetic resonance imaging (MRI) at 3 T followed by single-voxel MRS at three regions of interest (ROI) within MC/MTIII (lateral condyle, medial condyle, proximal bone marrow [PBM]). Percentage FC was calculated at each ROI. Each limb underwent computed tomography (CT) and bone mineral density (BMD) was calculated for the same ROIs. All MR and CT images were graded for sclerosis. Histology slides were graded for sclerosis and proximal marrow space was calculated. Pearson or Spearman correlations were used to assess the relationship between BMD, FC and marrow space. Kruskal-Wallis tests were used to check for differences between sclerosis groups for BMD or FC. Eighteen limbs from 10 horses were included. A negative correlation was identified for mean BMD and FC for the lateral condyle (correlation coefficient = -0.60, p = 0.01) and PBM (correlation coefficient = -0.5, p = 0.04). There was a significant difference between median BMD for different sclerosis grades in the condyles on both MRI and CT. A significant difference in FC was identified between sclerosis groups in the lateral condyle on MRI and CT. Small sample size. 1H Proton MRS is feasible in the equine MC/MTIII. Further work is required to evaluate the use of this technique to predict fracture risk in racehorses.

Abstract 1148: Validating Artificial Intelligence-assisted Classification Of Peripheral Arterial Disease Lesion Composition From 9.4 Tesla Magnetic Resonance Imaging With Histology

Objectives: MRI is a promising tool for identifying tissue properties and enhancing the precision of endovascular peripheral artery disease (PAD) treatment. However, the interpretation of images might pose a challenge for untrained experts. Hypothesis: Unsupervised artificial intelligence (AI) algorithms can assist in interpreting MRI images and classifying lesion components. Aims: To confirm the feasibility of a custom-made variational autoencoder (VAE) in evaluating lesion crossability through histological validation. Methods: 9.4T MRI images (150-micron isotropic resolution) were acquired on 20 chronic total occlusion PAD lesions using T2-weighted (T2w) and Ultrashort Echo Time (UTE) contrasts. Pseudo-colors (T2w: red, UTE: green) were created to enhance tissue visualization (Fig1). A VAE algorithm classified axial images into categories based on the lumen being occluded with hard or soft tissue. Lesions underwent histologic analysis, slides were matched with the corresponding VAE output images. Two readers assessed lesion crossability (non-crossable if >50% of lumen was occluded with hard tissue) on both VAE-generated images and histologic slides. Results: MRI produced 4013 psuedo-color images, and VAE successfully separated axial sections based on plaque composition. Histologic analysis yielded 91 cross-sections. The consensus reading identified 34% (n=31) of VAE images and 25% (n=23) of histologic slides as non-crossable, with a significant association between the histology-based and AI-based crossability (p<.001). Inter-rater reliability was good for both VAE-based and histology-based crossability assessments (intralass correlation coefficient (ICC)= 0.76 and 0.80, respectively). Conclusion: Our AI algorithm demonstrated a strong association with histologic validation, suggesting potential for enhancing image interpretation, and possibly contributing to improved planning for endovascular interventions.

ARGININE VASOPRESSIN DEFICIENCY INDUCES DECREASED CARDIAC FIBROSIS RATE AND MEAN ARTERIAL BLOOD PRESSURE IN RATS WITH EXPERIMENTAL ABDOMINAL AORTIC STENOSIS

Cardiac fibrosis (CF) involves the accumulation of extracellular matrix proteins in the interstitial space, often seen in various cardiac pathologies. Each year, cardiovascular diseases claim the lives of 17.9 million people worldwide. Cardiac fibroblasts and myofibroblasts, equipped with vasopressin receptors (AVP), play a crucial role in regulating cell signaling pathways associated with CF development. AVP, an immunomodulatory hormone, influences immune responses, and its deficiency results in a decrease in immune activity. Abdominal aortic stenosis (AS) serves as an experimental model for left cardiac overload hypertension and CF. Neurointermediate pituitary lobectomy (NIL) induces a permanent decrease in circulating AVP levels, showcasing regression of hepatic fibrosis in liver damage models. The quest for effective alternative treatments for cardiac fibrosis poses a substantial challenge for researchers. In this experiment, groups of male Wistar rats weighing approximately 250 g (8/group) were categorized into: 1) Intact Control (IC), 2) Abdominal Aortic Stenosis (AS), 3) Neurointermediate pituitary lobectomy (NIL), and 4) AS+NIL. The AS and AS+NIL groups underwent abdominal aortic stenosis surgery on day one of the experiment. By week 5, the NIL and NIL+AS groups underwent NIL surgery. Prior to sacrifice at week 10, intracarotid mean blood pressure (MBP) was assessed in all groups. Following sacrifice, the hearts were removed, fixed in neutral formalin, and processed in paraffn for histopathological examination using H-E, Masson's trichrome, and Sirius red staining. Results revealed an elevated MBP in the AS group (156 ± 27.1 vs. 114 ± 30 mmHg in the IC), normalized in the AS+NIL group (110 ± 40.9 mmHg), and a mild decrease in the NIL group (87 ± 7.6 mmHg). The evaluation of CF in histological slides concurred with the MBP findings; a significant increase in CF occurred in the AS group compared to the IC, NIL, and AS+NIL groups, with almost complete reversion of CF observed in the AS+NIL group. In summary, this study demonstrates that AVP deficiency induces arterial hypotension, and in the AS+NIL group, AVP deficiency led to: 1) a decrease in blood pressure to normal levels and 2) reversion of cardiac fibrosis. Further experiments are warranted to determine if the decrease in blood pressure and cardiac overload are suffcient to inhibit profibrogenic mechanisms and/or activate antifibrogenic mechanisms. Additionally, investigations are needed to explore whether AVP deficiency, beyond inhibiting the inflammatory process, promotes the activation of antifibrogenic mechanisms. The present results offer new avenues for a deeper understanding of the molecular and cellular mechanisms through which AVP deficiency or the use of AVP receptor antagonists significantly influences fibrosis regulation in organ fibrotic diseases. Funding by Autonomous University of Aguascalientes (PIFF22-1). This is the full abstract presented at the American Physiology Summit 2024 meeting and is only available in HTML format. There are no additional versions or additional content available for this abstract. Physiology was not involved in the peer review process.

Histomorphological changes in gills and hepatopancreas of crabs: an integrative review

Crustaceans are a biological model of ecotoxicology and consequently are widely used to assess the impact of xenobiotics. The toxicity of xenobiotics starts in the interaction of biomolecules and can even cause changes in an entire ecosystem. This paper aims to compile the histomorphological changes caused by xenobiotics in the gills and hepatopancreas of crabs through an integrative review. The literature search was done using PubMed with keywords in English and the filters "full text" and "publications from 2001-21". A total of 299 studies were found and 35 of them were selected based on the title and abstract. Later, 17 were included in the study. There were 18 categories of histomorphological changes in the gills and 22 in the hepatopancreas. We noticed that there is no consensus on the terminology used for histomorphological changes in this group. Furthermore, it was observed that the quality of the histological slides was quite diverse, making the identification and comparison of changes in the images problematic. The aforementioned considerations justify the need for this study to address a gap and encourage future research to develop a guide of histomorphological changes in crabs and its use as a tool for environmental biomonitoring.

The degradation of collagen in submerged bones, analysed by ImageJ® and Orbit®, for the estimation of the post-mortem interval (PMI) and the post-mortem submersion interval (PMSI)

ABSTRACT The estimation of the post-mortem submersion interval (PMSI) of skeletal remains can provide critical information to forensic investigations, contributing to various aspects such as identification. A fundamental aspect of the PMSI estimation in bone is the degradation of collagen, the most abundant organic component. This degradation can be assessed using various analytical techniques, including the digital analysis of images taken from histological slides. Two open-source software for image digital analysis, ImageJ® and Orbit®, provide novice-friendly and accessible technology to a broad audience of forensic practitioners. However, the absence of established procedures can challenge the consistency of the interpretation of the results, especially between observers. The principal aim of this study was to compare the efficacy of ImageJ® and Orbit® in quantifying the degradation of collagen in images taken from histological slides of mammalian non-human bones submerged for 4 months to 2 years. Despite the inconsistency in the outcomes of intraobserver vs. interobserver testing, this research has demonstrated the promising potential of digital histology. It has also identified the need for further investigation, particularly focusing on the usability of ImageJ® and the refinement of the Orbit® workflow.

Development of an Artificial-Intelligence-Based Tool for Automated Assessment of Cellularity in Bone Marrow Biopsies in Ph-Negative Myeloproliferative Neoplasms.

The cellularity assessment in bone marrow biopsies (BMBs) for the diagnosis of Philadelphia chromosome (Ph)-negative myeloproliferative neoplasms (MPNs) is a key diagnostic feature and is usually performed by the human eyes through an optical microscope with consequent inter-observer and intra-observer variability. Thus, the use of an automated tool may reduce variability, improving the uniformity of the evaluation. The aim of this work is to develop an accurate AI-based tool for the automated quantification of cellularity in BMB histology. A total of 55 BMB histological slides, diagnosed as Ph- MPN between January 2018 and June 2023 from the archives of the Pathology Unit of University "Luigi Vanvitelli" in Naples (Italy), were scanned on Ventana DP200 or Epredia P1000 and exported as whole-slide images (WSIs). Fifteen BMBs were randomly selected to obtain a training set of AI-based tools. An expert pathologist and a trained resident performed annotations of hematopoietic tissue and adipose tissue, and annotations were exported as .tiff images and .png labels with two colors (black for hematopoietic tissue and yellow for adipose tissue). Subsequently, we developed a semantic segmentation model for hematopoietic tissue and adipose tissue. The remaining 40 BMBs were used for model verification. The performance of our model was compared with an evaluation of the cellularity of five expert hematopathologists and three trainees; we obtained an optimal concordance between our model and the expert pathologists' evaluation, with poorer concordance for trainees. There were no significant differences in cellularity assessments between two different scanners.