Bradley R, Tagkopoulos I, Kim M, Kokkinos Y, Panagiotakos T, Kennedy J, De Meyer G, Watson P, Elliott J. Predicting early risk of chronic kidney disease in cats using routine clinical laboratory tests and machine learning. J Vet Intern Med. 2019; doi:10.1111/jvim.15623.
This study aimed to develop a model predicting chronic kidney disease (CKD) in cats using electronic health records (EHRs) from 106,251 cats treated at Banfield Pet Hospitals. A recurrent neural network (RNN) model was created, incorporating creatinine, blood urea nitrogen, urine specific gravity, and age as predictive features. The model showed 90.7% sensitivity and 98.9% specificity near diagnosis, with sensitivity decreasing as prediction windows lengthened (63.0% at 1 year and 44.2% at 2 years prior). Machine learning models can enhance early detection of CKD, aiding veterinary decision-making.
Comments