The digital transformation of pathology, through automation and computational tools, addresses current challenges in the field. This study evaluates Paige Pan Cancer, a novel artificial intelligence tool based on the Virchow foundation model, designed to flag invasive cancer in haematoxylin and eosin-stained slides from 16 primary tissue types. Using 62 cases from the Ipatimup Pathology Laboratory, we found the tool had a sensitivity of 93.3% and specificity of 87.5% in biopsies, and 94.7% sensitivity and 75.0% specificity in resections. Overall accuracy was 90.3%. Despite some misclassifications, Paige Pan Cancer demonstrates high sensitivity as a multi-organ screening tool in clinical practice.
Keywords: artificial intelligence; cancer diagnosis; computational pathology; efficiency; foundation model; workflow.
© 2024. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.