A histopathology image dataset for Johne’s disease detection and research

Abstract

Johne’s disease (paratuberculosis), caused by Mycobacterium avium subspecies paratuberculosis, is a chronic intestinal infection that affects ruminants and poses significant challenges for livestock health and management. Accurate and early diagnosis is of paramount importance for effective disease control, yet traditional histopathological assessment requires expert interpretation and remains subject to interobserver variability. In this paper, we present a curated dataset of histopathological slide images collected from tissue samples confirmed to be positive or negative for Johne’s disease. The samples were processed and stained using standard hematoxylin and eosin (H&E) protocols, and the slides were digitized using a high-resolution microscope camera. Each image is annotated with a diagnostic label verified by a board-certified pathologist. The dataset is organized by disease status (i.e., positive vs. negative), which makes it useful for supervised machine learning applications, computer-aided diagnosis, and digital pathology research. In addition to supporting the development of automated detection systems, the dataset serves as a valuable educational resource for training veterinary pathology students in recognizing histological patterns associated with MAP infection. To the best of our knowledge, this is the first publicly available dataset of histopathology images dedicated to Johne’s disease.