Inter-Rater and Intra-Rater Agreement in Scoring Severity of Rodent Cardiomyopathy and Relation to Artificial Intelligence-Based Scoring.

We previously developed a computer-assisted image analysis algorithm to detect and quantify the microscopic features of rodent progressive cardiomyopathy (PCM) in rat heart histologic sections and validated the results with a panel of five veterinary toxicologic pathologists using a multinomial logistic model. In this study, we assessed both the inter-rater and intra-rater agreement of the pathologists and compared pathologists' ratings to the artificial intelligence (AI)-predicted scores. Pathologists and the AI algorithm were presented with 500 slides of rodent heart. They quantified the amount of cardiomyopathy in each slide. A total of 200 of these slides were novel to this study, whereas 100 slides were intentionally selected for repetition from the previous study. After a washout period of more than six months, the repeated slides were examined to assess intra-rater agreement among pathologists. We found the intra-rater agreement to be substantial, with weighted Cohen's kappa values ranging from k = 0.64 to 0.80. Intra-rater variability is not a concern for the deterministic AI. The inter-rater agreement across pathologists was moderate (Cohen's kappa k = 0.56). These results demonstrate the utility of AI algorithms as a tool for pathologists to increase sensitivity and specificity for the histopathologic assessment of the heart in toxicology studies.

Using Artificial Intelligence to Detect, Classify, and Objectively Score Severity of Rodent Cardiomyopathy.

Rodent progressive cardiomyopathy (PCM) encompasses a constellation of microscopic findings commonly seen as a spontaneous background change in rat and mouse hearts. Primary histologic features of PCM include varying degrees of cardiomyocyte degeneration/necrosis, mononuclear cell infiltration, and fibrosis. Mineralization can also occur. Cardiotoxicity may increase the incidence and severity of PCM, and toxicity-related morphologic changes can overlap with those of PCM. Consequently, sensitive and consistent detection and quantification of PCM features are needed to help differentiate spontaneous from test article-related findings. To address this, we developed a computer-assisted image analysis algorithm, facilitated by a fully convolutional network deep learning technique, to detect and quantify the microscopic features of PCM (degeneration/necrosis, fibrosis, mononuclear cell infiltration, mineralization) in rat heart histologic sections. The trained algorithm achieved high values for accuracy, intersection over union, and dice coefficient for each feature. Further, there was a strong positive correlation between the percentage area of the heart predicted to have PCM lesions by the algorithm and the median severity grade assigned by a panel of veterinary toxicologic pathologists following light microscopic evaluation. By providing objective and sensitive quantification of the microscopic features of PCM, deep learning algorithms could assist pathologists in discerning cardiotoxicity-associated changes.

The National Toxicology Program Web-based nonneoplastic lesion atlas: a global toxicology and pathology resource.

Toxicologists and pathologists worldwide will benefit from a new, website-based, and completely searchable Nonneoplastic Lesion Atlas just released by the U.S. National Toxicology Program (NTP). The atlas is a much-needed resource with thousands of high-quality, zoomable images and diagnostic guidelines for each rodent lesion. Liver, gallbladder, nervous system, bone marrow, lower urinary tract and skin lesion images, and diagnostic strategies are available now. More organ and biological systems will be added with a total of 22 chapters planned for the completed project. The atlas will be used by the NTP and its many pathology partners to standardize lesion diagnosis, terminology, and the way lesions are recorded. The goal is to improve our understanding of nonneoplastic lesions and the consistency and accuracy of their diagnosis between pathologists and laboratories. The atlas is also a useful training tool for pathology residents and can be used to bolster any organization's own lesion databases. Researchers have free access to this online resource at www.ntp.niehs.nih.gov/nonneoplastic.

Immunohistochemical characterization of spontaneous and acrylonitrile-induced brain tumors in the rat.

Twenty-eight spontaneously occurring glial tumors (previously diagnosed as astrocytomas, oligodendrogliomas, and gliomas) and eleven granular cell tumors (GCTs) were selected for evaluation using a panel of immunohistochemistry (IHC) stains (Ricinus communis agglutinin type 1 [RCA-1], ionized calcium-binding adapter molecule 1 [Iba-1], OX-6/major immunohistocompatibility complex class II, oligodendrocytes transcription factor 2 [Olig2], glial fibrillary acidic protein [GFAP], S100 beta, glutamine synthetase, neurofilament, proliferating cell nuclear antigen). In addition, nine brain tumors from a 2-year drinking water study for acrylonitrile were obtained from the Acrylonitrile Group, Inc. Based on IHC staining characteristics, Olig2+ oligodendrogliomas were the most commonly diagnosed spontaneous tumor in these animals. Many of the spontaneous tumors previously diagnosed as astrocytomas were RCA-1+, Iba-1+ and negative for GFAP, S100beta, and glutamine synthetase; the diagnosis of malignant microglial tumor is proposed for these neoplasms. Three mixed tumors were identified with Olig2+ (oligodendrocytes) and Iba-1+ (macrophage/microglia) cell populations. The term mixed glioma is not recommended for these tumors, as it is generally used to refer to oligoastrocytomas, which were not observed in this study. GCT were positive for RCA-1 and Iba-1. All acrylonitrile tumors were identified as malignant microglial tumors. These results may indicate that oligodendrogliomas are more common as spontaneous tumors, while acrylonitrile-induced neoplasms are microglial/histiocytic in origin. No astrocytomas (GFAP, S100 beta, and/or glutamine synthetase-positive neoplasms) were observed.