o:3278 Automated Nuclear Morphometry: A Deep Learning Approach for Prognostication in Canine Pulmonary Carcinoma to Enhance Reproducibility en The integration of deep learning-based tools into diagnostic workflows is increasingly prevalent due to their efficiency and reproducibility in various settings. We investigated the utility of automated nuclear morphometry for assessing nuclear pleomorphism (NP), a criterion of malignancy in the current grading system in canine pulmonary carcinoma (cPC), and its prognostic implications. We developed a deep learning-based algorithm for evaluating NP (variation in size, i.e., anisokaryosis and/or shape) using a segmentation model. Its performance was evaluated on 46 cPC cases with comprehensive follow-up data regarding its accuracy in nuclear segmentation and its prognostic ability. Its assessment of NP was compared to manual morphometry and established prognostic tests (pathologists' NP estimates (n = 11), mitotic count, histological grading, and TNM-stage). The standard deviation (SD) of the nuclear area, indicative of anisokaryosis, exhibited good discriminatory ability for tumor-specific survival, with an area under the curve (AUC) of 0.80 and a hazard ratio (HR) of 3.38. The algorithm achieved values comparable to manual morphometry. In contrast, the pathologists' estimates of anisokaryosis resulted in HR values ranging from 0.86 to 34.8, with slight inter-observer reproducibility (k = 0.204). Other conventional tests had no significant prognostic value in our study cohort. Fully automated morphometry promises a time-efficient and reproducible assessment of NP with a high prognostic value. Further refinement of the algorithm, particularly to address undersegmentation, and application to a larger study population are required. Primary Lung-Tumors; Prostate-Cancer; Dogs; Classification; Survival; Stereology; Prognosis; Neoplasia; Biopsies 1552099 10.3390/vetsci11060278 2024-07-24T07:19:23.372Z 44 yes 46 Imaine Glahn University of Veterinary Medicine Vienna 0009-0003-9247-9302 Christof A. Bertram University of Veterinary Medicine Vienna person 0000-0002-2402-9997 Stephan M. Winkler University of Applied Sciences Upper Austria / Johannes Kepler University of Linz person Marc Aubreville Technische Hochschule Ingolstadt person 0000-0002-5294-5247 Josef Scharinger Johannes Kepler University of Linz person 0000-0001-6502-7501 Jonathan Ganz Technische Hochschule Ingolstadt person F. Yvonne Schulman Mars Petcare Science and Diagnostics person 0000-0002-7454-8517 Barbara Richter University of Veterinary Medicine Vienna person Sophie Merz IDEXX Vet Med Labor GmbH person Robert Klopfleisch Freie Universität Berlin person 0000-0002-6308-0568 Andrea Klang University of Veterinary Medicine Vienna person Michael J. Dark University of Florida person 0000-0002-6547-1298 Pompei Bolfa Ross University School of Veterinary Medicine person 0000-0002-2903-1535 Florian Bartenschlager Freie Universität Berlin person Charles-Antoine Assenmacher University of Pennsylvania person 0000-0003-0073-2730 Hannah Janout University of Applied Sciences Upper Austria / Johannes Kepler University of Linz person 0000-0002-0294-3585 Philip S. Hyndman The Schwarzman Animal Medical Center person 0000-0003-4367-4032 Theresa Kreilmeier-Berger University of Veterinary Medicine Vienna person 0000-0001-5346-2821 Alexander Bartel Freie Universität Berlin person 0000-0002-1280-6138 Brigitte Degasperi University of Veterinary Medicine Vienna person 0009-0001-0559-3728 Andreas Haghofer University of Applied Sciences Upper Austria / Johannes Kepler University of Linz person 0000-0001-6649-5374 Taryn A. Donovan The Schwarzman Animal Medical Center person 0000-0001-5740-9550 application/pdf 5907061 https://phaidra.vetmeduni.ac.at/o:3278 no yes 16 70 1552253 Veterinary Sciences 21 11 6 MDPI 2024