The coiled-coil domain-containing 85c (Ccdc85c) knockout (KO) rat generated by genome editing exhibits hydrocephalus and subcortical heterotopia. In this study, we aimed to further investigate the function of CCDC85C protein in the development of the retina. Expression of CCDC85C, acetylated tubulin, ciliary rootlet coiled-coil protein (CROCC), zonula occludens-1 (ZO-1), glutamine synthetase, and PAX6 were examined immunohistochemically in wild-type F344 rats at embryonic day (ED) 19 and at postnatal days (PNDs) 0, 4, 6, 13, and 20. Immunoelectron microscopy was performed for CCDC85C in the normal rat retina. Retinal lesions in Ccdc85c KO rats were examined using fundus photography, optical coherence tomography (OCT), and histology. In the normal rat retina, CCDC85C was co-localized with ZO-1 in the outer limiting membrane and persistently expressed after ED19. Ultrastructurally, CCDC85C was located between the outer nuclear layer and the inner segments, and showed the same location as the tight junction. In Ccdc85c KO rats, multifocal retinal dysplasia; disarrangement of the inner and outer segments, cilia, and rootlets; and impaired development of Müller cells were observed. In OCT images, Ccdc85c KO rats showed parallel hyperintense striations in the inner nuclear layer, and low reflectivity of the outer limiting membrane and layer of rods and cones. These results suggest that CCDC85C protein is located in the tight junction complex and is involved in retinal layer formation. The Ccdc85c KO rat model provides a novel tool to study retinal development as well as genetic hydrocephalus.
The advent of artificial intelligence (AI) technologies is creating a paradigm shift in drug discovery and development. Veterinary pathology is an area that can significantly benefit from AI tools. The availability of AI algorithms in commercial software packages such as HALO and Visiopharm has generated interest in automating pathologists' workflows for the detection and quantification of histology endpoints from whole-slide images. As each software package provides a distinct set of AI models, the relative performance and equivalency of these models for lesion detection and quantification are poorly understood. Here, we present a systematic comparison of the performance of AI algorithms developed in HALO and Visiopharm. Specifically, we trained AI algorithms in HALO and Visiopharm by using the same ground truth training data to detect different subregions and quantify their areas from hematoxylin and eosin (HE)-stained images of murine skin sections. We also calculate the performance metrics (precision, recall, F1 score) for each algorithm using the same test dataset. Our analysis shows that both HALO and Visiopharm algorithms have comparable performance and are resilient to training data size and changes in the color profile of the HE images. As an application, we compared the results of the AI algorithms against pathologists' scores to quantify epidermal hyperplasia. Our analysis shows that the quantitative data from AI algorithms are consistent with pathologists' scores. These results provide a quantitative characterization of commercially available AI models in HALO and Visiopharm and offer practical guidelines for designing and validating AI algorithms using these software packages.
The efficacy of candidate drugs for neuroprotection is tested in rodent models of retinal atrophy where light-induced outer retinal atrophy (ORA) is quantified by manual or semi-automated measurements, analyses that are time-consuming and error-prone. We developed a quantitative, automated image analysis-based method of ORA assessment in whole-slide images (WSIs). A commercial, cloud-based, artificial intelligence image analysis platform (Aiforia) was used to train convolutional neural network (CNN)-based deep learning models for semantic segmentation of retina and object counts of outer nuclear layer (ONL) nuclei. Model development was an iterative process of establishing and fine-tuning the ground truth (manual annotations), adding training annotations, and increasing the number of CNN training iterations. Inter-observer concordance yielded F1-scores exceeding 0.98 for retina area and 0.92 for ONL counts. Performance of the developed algorithm in comparison with manual annotations by individual validators yielded F1-scores between 0.89 and 0.95. Analysis of WSIs by the algorithm was completed 50× faster compared with manual analysis on hematoxylin and eosin-stained sections; however, it required time to digitize the slides. Performance of the ORA quantification deep learning model is noninferior to that of current manual ORA scoring approaches and provides improved reproducibility and reduced time and cost of analysis.
Lymphoma is a common cancer in dogs, which often presents as generalized peripheral lymphadenopathy. Fine needle aspiration is frequently used to investigate the cause of peripheral lymphadenopathy, and artificial intelligence technology could potentially assist in cytologic interpretation. In this study, YOLOv11, an open-source object detection algorithm, was evaluated for cell-level identification in canine lymph node cytology to diagnose lymphoma. Cytologic images were captured using 2 smartphones from 11 non-lymphoma and 34 intermediate-to-large B-cell lymphoma peripheral lymph node aspirates in dogs, confirmed by cytologic interpretation plus either flow cytometry or polymerase chain reaction for antigen receptor rearrangements. A total of 25,761 intact cells were annotated across 680 images. Models were trained and validated under 5 cross-device configurations, comparing a 4-label approach (small, intermediate, and large lymphocytes and neutrophils) and a 3-label approach that merged intermediate and large lymphocytes. At the cell level, combining intermediate and large lymphocytes improved cell-level classification. In the optimized configuration, the model achieved an average precision of 87.60%, an average recall of 84.71%, a mean average precision 50 (mAP50) of 89.33%, and an F1 score of 86.02% on the independent test set. At the aspirate level, the model achieved near-perfect performance and correctly inferred diagnoses in all test aspirates by assessing the proportions of intermediate-to-large lymphocytes. These findings demonstrate the potential feasibility of smartphone-based deep learning assistance for veterinary cytology and highlight the importance of prospective, workflow-integrated validation, including external validation, across a broader range of lymph node diseases.
Hemorrhagic necrotizing pneumonia caused by cytotoxic necrotizing factor-1-positive extraintestinal pathogenic Escherichia coli (ExPEC) is a rapidly fatal disease in dogs. Lung tissues from 6 client-owned dogs with culture-confirmed ExPEC pneumonia were analyzed using immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH). Both assays localized E. coli within pneumocytes, alveolar macrophages, and alveolar spaces. Notably, one case exhibited numerous filamentous E. coli, representing a stress-induced morphologic adaptation previously unreported in canine ExPEC pneumonia. Recognizing filamentous E. coli is diagnostically crucial to avoid misidentification as other Gram-negative bacteria. While FISH provided high specificity, IHC offered practical, rapid detection compatible with routine microscopy. Four dogs had recent exposures to high-density facilities (breeding, boarding, or training), suggesting possible environmental transmission. These findings expand the clinicopathologic spectrum of canine ExPEC pneumonia and demonstrate the value of integrating spatial techniques with traditional culture to improve diagnostic accuracy.
Dental lesions are often painful and can impair digestive efficiency and fitness in mammals, consequently leading to systemic health complications. Although dental lesions are rarely described in wild herbivores, a high prevalence of incisor abnormalities was detected in wild muskoxen (Ovibos moschatus) using mandibles collected by Inuit hunters through the Community-Based Wildlife Health Surveillance program in the Canadian Arctic. Here, we present the first comprehensive examination of incisor dental lesions and developmental anomalies in wild muskoxen. We examined 243 mandibles from harvested muskoxen and identified and described both gross and radiographic pathology. The most frequent lesion was enamel infractions, characterized by linear cracks along lingual enamel surfaces, which was present in 91.4% (222/243) of individuals. Incisor fractures, which featured at least one incisor with a complicated crown fracture, were present in 50.6% (123/243) of individuals. Labial enamel abrasions were present in 26.3% (64/243) of muskoxen in which wear extended to the dentin, and in severe cases to the pulp. Other common gross abnormalities included rotation (31.3%; 76/243), crowding (21.8%; 53/243), and unerupted fourth incisors (18.9%; 46/243). Radiographic analysis revealed congenitally absent teeth, root fractures, and endodontic-related lesions that correlated with gross incisor abnormalities in a subset of 136 animals. Other radiographic findings included hypercementosis, ankylosis, dilaceration, dental decay, and retained primary teeth. These results underscore the high prevalence and severity of incisor pathology in muskoxen and provide foundational data for subsequent assessment of lesion epidemiology and their association with muskox health.
Congenital malformations (CMFs) are structural disorders that arise during embryogenesis, and various forms of these abnormalities have been reported in reptiles. The yacare caiman ( Caiman yacare ) is widely used in commercial farms in the Pantanal region, and studies characterizing CMFs in these animals are scarce. This study describes the pathological and imaging findings of CMFs in farmed caimans. From February 2022 to February 2023, the incubation and maternity of a caiman commercial farm, operating in ranching and farming systems, were monitored. A total of 26 distinct CMFs were identified in 30 farmed caimans. All animals were classified as infantile (9.65 ± 2.69 cm snout-vent length). CMFs were grouped according to anatomical site, with craniofacial disorders (18 cases) being the most frequent findings, followed by spinal malformations (13 cases). Palatoschisis (PC) and cheiloschisis (CL) were reported in 6 yacare caimans, which also exhibited a malformed palatal velum and secondary aspiration pneumonia. Meningoencephalocele was diagnosed in 3 caimans, with brain hemorrhage being a typical secondary lesion in these cases. Anophthalmus, exophthalmos, and cyclopia represented the ocular CMFs. Spinal malformations (kyphosis and scoliosis) were detected in 13 yacare caimans and were thoroughly characterized by the X-ray radiographic analysis. Tail abnormalities were reported in 12 animals, including tail agenesia (1 case) and coiled tail (11 cases). Other sites of CMFs include the limbs, the coelomic wall, and the skin. Radiographic assessments were obtained from all 30 CMFs, providing detailed information on these lesions.
A distinctive peritumoral epithelial proliferation was identified exclusively in the labial region of Labrador retrievers in association with sebaceous neoplasms. To characterize this lesion, 515 sebaceous neoplasms diagnosed in dogs of different breeds between 2017 and 2025 were retrospectively reviewed. Among these, 2 of 255 epitheliomas (0.7%) and 22 of 41 adenocarcinomas (54%) arising in the lips exhibited a distinctive peripheral basaloid epithelial proliferation. The lesion was observed only in the lower lip of Labrador retrievers, occurring in association with either solitary or multifocal neoplastic nodules. No similar proliferations were identified after examining 103 labial lesions of other histotypes diagnosed in the same timeframe. Histologically, the proliferations consisted of 1- to 2-cell-thick epithelial cords surrounding pre-existing adnexal units. The cells showed minimal atypia and low mitotic activity. Immunohistochemically, they exhibited a cytokeratin (CK)5+/p63+/alpha-smooth muscle actin-/CK7- phenotype, supporting derivation from basal reserve cells of cutaneous glands and excluding differentiation toward mature apocrine epithelial or myoepithelial cells. The exclusive occurrence in Labrador retrievers suggests a possible breed-related predisposition. Although the biological behavior of this novel lesion is unknown, its recognition may be important in diagnostic pathology to avoid misinterpretation with an invasive adenocarcinoma.
Feline gammaherpesvirus 1 (FcaGHV1) is a recently identified member of the Gammaherpesvirinae detected in domestic cats; however, its tissue distribution and pathological relevance remain incompletely defined. This study investigated the localization of FcaGHV1 in naturally infected cats and evaluated potential relationships between viral presence and histopathologic lesions. Fresh tissues from 42 domestic cats submitted for routine necropsy in Thailand were screened for FcaGHV1 using quantitative polymerase chain reaction targeting the glycoprotein B gene, followed by sequencing confirmation. Corresponding formalin-fixed tissues were examined histologically, and viral localization was assessed using chromogenic in situ hybridization (ISH). FcaGHV1 was detected in 4 of 42 cats (10%). Viral loads were highest in lymph nodes and spleens, and ISH localized viral nucleic acids to histiocytes and small lymphocytes within lymphoid tissues. In 2 infected cats, pulmonary lesions characterized by interstitial fibrosis and vascular remodeling were observed, and viral hybridization signals were detected within infiltrating mononuclear cells, spindle-shaped interstitial cells, and vascular-associated stromal cells in affected regions. Based on this observation, an additional retrospective cohort of 18 cats with histologically diagnosed pulmonary fibrosis was examined using ISH. Viral hybridization signals were detected in 3 of 18 lungs (17%) and were primarily localized within fibrotic areas. Together, these findings demonstrate tissue-level evidence of FcaGHV1 localization in lymphoid organs and pulmonary lesions of naturally infected cats. The repeated detection of viral nucleic acids within fibrotic lung lesions across independent cohorts suggests an association between FcaGHV1 localization and pulmonary remodeling; however, the biological significance of this observation remains uncertain.
Cognitive biases are systematic patterns of error in human thinking. While artificial intelligence (AI)-based decision support systems offer great potential to enhance diagnostic accuracy and efficiency in pathology, collaboration between medical professionals and AI can also introduce or amplify these biases. This review aims to identify cognitive biases and their modes of manifestation in human-computer interaction (HCI) within pathology and to extrapolate potential manifestations of biases not yet explored in this domain but reported in HCIs across other medical specialties. We conducted a structured literature review (19 August 2025) across the ACM, IEEE, and PubMed databases. Studies were eligible if they operationalized cognitive biases during expert-machine interaction in diagnostic decision-making using qualitative or quantitative methods, including review articles citing primary studies that met these criteria. The final corpus comprised 24 studies (8 primary studies and 16 review articles). From these review articles, 18 additional primary studies were extracted and used in their place solely for analysis purposes. A narrative synthesis of the identified primary research revealed 12 cognitive biases reported in AI-assisted medical decision-making, with only one study originating from pathology. For each bias, definitions and hypothetical pathology-specific examples were derived. This review is intended as a primer for the veterinary and human pathology communities and strives to contribute to the safe and effective integration of AI into diagnostic practice.
While pleomorphic adenoma (PA) is the most common salivary neoplasm in human medicine, it appears to be rare in animals and has only recently been reported in a rhesus macaque. We gathered 6 cases of PAs affecting aged rhesus macaques and summarized their gross, histological, and immunohistochemical characteristics. Affected animals ranged from 19 to 32 years, were predominantly male, and the neoplasm was an incidental finding at necropsy. The neoplasms affected the submandibular salivary gland (5/6) and the parotid gland (1/6). Neoplasms were round to ovoid, tan, well-demarcated, and firm. Histologically, the masses had an epithelial and a myoepithelial component supported by a hyalinized fibrous stroma. Immunohistochemistry (IHC) was performed in 5 cases. Pancytokeratin expression was diffuse in the epithelial component. Pleomorphic adenoma gene 1 (PLAG1), a specific nuclear IHC marker for human PAs, was negative in all but 1 case, where rare myoepithelial cells had false-positive cytoplasmic immunolabeling. Calponin and smooth muscle actin (SMA) highlighted the myoepithelial component, with calponin showing stronger expression than SMA in both the number and distribution of immunolabeled myoepithelial cells. Three cases had less than or equal to 5% of cells expressing Ki-67; in 2 cases that were larger with areas of necrosis, expression ranged from 10% to 20% of cells. In summary, salivary PAs in rhesus macaques are usually incidental and apparently affect mainly the submandibular salivary gland in aged males. Immunohistochemistry for PLAG1 is not helpful in the diagnosis, which instead relies primarily on gross and histological findings.
In 2023, high pathogenicity avian influenza (HPAI) H5N1 outbreaks caused mass mortality events affecting pinnipeds throughout South America. Here, we present the clinical, pathological, immunohistochemical, and molecular findings associated with HPAI H5N1 infection in 3 South American sea lions (Otaria flavescens) and 1 southern elephant seal (Mirounga leonina) stranded during the outbreak in Chubut, Argentina. Neurological signs were observed in 2 sea lions, with 1 also exhibiting respiratory distress. Necropsies were conducted on all 4 animals. Neuropathology revealed mild-to-severe lymphohistiocytic (2/4) to mixed neutrophilic (2/4) meningoencephalitis with neuronal necrosis, neuronophagia, glial cell proliferation, multifocal hemorrhage, and perivascular cuffing in all animals. Novel findings for HPAI H5N1-infected pinnipeds included choroid plexitis (2/4) and myelitis (2/4). Immunohistochemistry for viral nucleoprotein was positive in 3 of 4 animals and involved neurons (3/3), glial cells (3/3), and ependymal cells (2/3). Viral-related systemic findings included multifocal necrotizing myocarditis in the elephant seal and multifocal necrotizing placentitis in a sea lion, with HPAI virus detected within fetal tissues. Pulmonary lesions were minimal and were limited to multifocal necrosis of bronchial glands in 1 sea lion. HPAI H5 clade 2.3.4.4b virus was confirmed by polymerase chain reaction and sequencing in all animals. These findings indicate a predominance of central nervous system involvement in HPAI H5N1-infected pinnipeds and expand the recognized spectrum of virus-associated lesions, identifying novel tissue tropisms and transmission routes.
Blackleg is a necrotizing, hemorrhagic, and emphysematous myositis of ruminants caused by Clostridium chauvoei. Here, we report a previously unrecognized manifestation of blackleg in calves characterized by multifocal, transmural intestinal necrosis. Two, 3- and 6-month-old beef calves without history of vaccination against clostridial diseases died following acute muscular disease. Gross and histologic changes were consistent with blackleg, including emphysematous and necrotizing myositis with gram-positive bacilli. Both calves had sharply demarcated areas of transmural, intestinal coagulative necrosis with vascular congestion, hemorrhage, fibrinoid necrosis, and thrombosis. Infection was confirmed by positive C. chauvoei immunolabeling in skeletal muscles of both calves and the small intestine of one case, and by polymerase chain reaction (PCR) in the skeletal muscles and intestines of both calves. Intestinal necrosis has not been previously reported in association with blackleg. Systematic examination, including the intestines, in cases of blackleg is recommended.
Q fever, caused by Coxiella burnetii, leads to debilitating human infections that are often recalcitrant to antibiotics. The development of new medical countermeasures is critical, as the only available vaccine has significant limitations that prevent its widespread or rapid use. To address this, we characterized aerosol C. burnetii infection in 16 cynomolgus macaques (Macaca fascicularis), providing the first detailed pathological description of Q fever pneumonia in a nonhuman primate model. Following exposure, all animals developed clinical signs consistent with human Q fever, including sustained fever, respiratory distress, and weight loss. Clinical pathology revealed hematological changes and evidence of systemic inflammation, including elevated C-reactive protein. Although clinical signs resolved by day 28, significant pathology persisted, including granulomatous interstitial pneumonia, pleuritis, carditis, hepatitis, and nephritis. Infection was confirmed in tissues using immunohistochemistry and electron microscopy. The cynomolgus macaque model faithfully replicates key features of human Q fever, validating its utility for studying disease pathogenesis and serving as a pivotal model for evaluating next-generation vaccines and therapies.
Veterinary electronic health records are often stored as unstructured free text, and structuring this information into analyzable formats is essential for downstream research. Natural language processing methods, including rule-based systems, machine learning algorithms, and, more recently, large language models (LLMs), provide tools to achieve this goal. This scoping review follows the PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) Extension for Scoping Reviews (PRISMA-ScR) guideline to examine how information extraction (IE) has been applied in veterinary medicine and extended to existing LLM approaches in human medicine. Literature research was conducted on 4 databases: PubMed, CAB Abstracts, Web of Science, and ACL Anthology, with stricter criteria limiting human medicine to prompt-based studies. After screening 5796 original research papers, a total of 23 veterinary and 31 human studies were selected for inclusion. In the veterinary literature, larger data sets were more commonly used to train supervised models, whereas human studies increasingly employed prompt-based LLMs, such as LLaMA and GPT, enabling IE with smaller annotated data sets. We developed a practical framework covering data preparation, platform and privacy considerations, and prompt engineering, with a corresponding workflow and prompt example to demonstrate the application of artificial intelligence (AI) in clinical data extraction. This review offers a practical tool to help veterinary researchers effectively integrate AI tools and LLMs into clinical research workflows.
Nonclinical toxicity studies are essential components of the drug development process and are based on labor-intensive visual examination of large numbers of glass slides under the microscope by pathologists. In this work, we present an artificial intelligence (AI)-based solution using an unsupervised representation learning model with a Bidirectional Generative Adversarial Network (BiGAN) for triaging normal and abnormal tissues and a machine learning-based severity grade classifier (SGC) for severity grade estimation on whole-slide images (WSIs) of rat liver. The BiGAN model was trained solely on vehicle control WSIs to learn normal histology and computed a tile-level similarity error, representing the deviation from normalcy, which was used to identify tile- and slide-level abnormalities. Our BiGAN model demonstrated an abnormality discrimination of receiver operating characteristic (ROC) area under the curve (AUC) of 0.77 with highest sensitivity for high-grade abnormalities. We utilized the similarity error data from BiGAN to train the SGC model and predicted slide-level severity grades with ROC AUC values ranging from 0.68 to 0.96 (minimal to marked) with most errors within ±1 grade, reflecting real-world pathologist grading variability. In addition, our heatmap visualizations of tile-level abnormalities revealed good to fair agreement with ground truth abnormalities in 70% of evaluated WSIs. By training the BiGAN on normal tissue slides, including minimal grade background abnormalities from vehicle control animals, our workflow eliminates annotation requirements and enables rapid adaptation to new tissues and species, paving the way for scalable AI models that streamline nonclinical pathology with minimal pathologist input.
In March and April of 2023, an outbreak of highly pathogenic avian influenza virus, H5N1 strain Eurasian lineage goose/Guangdong H5 clade 2.3.4.4b, resulted in at least 17 mortalities of free-ranging California condors (Gymnogyps californianus) in Arizona. Condors presented dead or with neurologic signs and lethargy. Infection resulted in multisystemic inflammation and necrosis that most consistently and severely affected the brain, spleen, and adrenal glands. Immunohistochemistry performed on tissues from a subset of condors labeled cells in multiple organ systems for influenza A virus, most abundantly in neurons, epithelial cells, and mononuclear inflammatory cells and often colocalized with areas of inflammation. Acute blunt force trauma, presumably from ground collision after falling from a height, was a common finding and indicated a rapidly debilitating disease course and neurologic impairment. Hepatic lead concentrations were relatively low with no concurrent incidences of acute lead toxicosis, an otherwise common cause of death in free-ranging California condors. Bone lead reflected long-term lead accumulation in several condors. Assessment of ingesta in 8 condors via morphologic hair identification showed a mix of consumed taxa, most commonly Bovidae. In summary, California condors, like other New World vultures (family Cathartidae), are highly susceptible to this strain of H5N1, and this should be taken into consideration when planning release, feeding, and morbidity and mortality responses.
