Sonographic machine-assisted recognition and tracking of B-lines in dogs

The SMARTDOG study evaluates an AI-based ultrasound algorithm for detecting B-lines in dogs with cardiogenic pulmonary edema (CPE). Results show strong agreement (ICC = 0.88) between AI and manual operator counts, with 84-86% accuracy in classifying pathological lung patterns. While AI failed in 14.2% of cases (more often in CPE dogs), it shows promise as a decision-support tool in veterinary critical care.