Wearable and contactless sensing technologies increasingly quantify animal movement, posture, activity, temperature, facial characteristics, vocal or respiratory events, and other observable signals potentially relevant to pain, lameness, stress, disease, and behavioral change. However, technical recognition performance, biological interpretation, external validation, and clinical actionability are frequently treated as if they represent the same evidential stage. This evidence-mapping review examined sensor modalities, species coverage, target outcomes, validation maturity, and clinical readiness in peer-reviewed veterinary and animal-health literature published from 2017 to 2026. The verified search identified 105 result appearances; 10 duplicate appearances were removed, leaving 95 unique records for screening. Forty-five reports underwent detailed eligibility assessment and 36 evidence units were included. The map comprised 26 primary empirical studies and 10 reviews or broader syntheses, with primary evidence concentrated in cattle. Wearable accelerometry, inertial sensing, computer vision, thermal imaging, bioacoustic monitoring, and multimodal systems addressed behavior, lameness, pain, health alerts, disease-related signs, and related states, but the evidence was uneven across species and endpoints. A proposed validation-maturity classification separates proof-of-concept and internal validation from independent-animal, external, multisite, and prospective implementation evidence; a complementary proposed clinical-readiness classification separates technical measurement from decision-support and demonstrated outcome utility. The evidence indicates that strong within-study classification does not by itself establish transportability, diagnosis, or clinical readiness. Major boundaries concern reference standards, individual-animal leakage, case-mix shift, environmental variability, measurement stability, and reliance on behavioral or physiological proxies. Future progress requires prospective multisite validation tied to clinically meaningful reference standards and explicit evaluation of how sensor-derived alerts alter veterinary decisions and outcomes.
