Machine learning for syndromic surveillance using veterinary necropsy reports.

Bollig N, Clarke L, Elsmo E, Craven M. Machine learning for syndromic surveillance using veterinary necropsy reports. PLoS One. Published February 5, 2020. https://doi.org/10.1371/journal.pone.0228105

This study explores the application of machine learning for syndromic surveillance using free-text veterinary necropsy reports. It focuses on detecting evidence of gastrointestinal, respiratory, and urinary pathologies in reports from the Wisconsin Veterinary Diagnostic Laboratory. Among various machine learning algorithms tested, the random forest model with TF-IDF feature vectors achieved high performance with F1 scores of 0.923 (gastrointestinal), 0.960 (respiratory), and 0.888 (urinary). The system analyzed over 33,000 necropsy reports over 14 years, revealing epidemiological trends and identifying a gastrointestinal disease cluster in late 2016 from a single producer.