A Narrative Review of Early Pregnancy Diagnosis Technologies for Livestock: From Conventional to Intelligent Systems

Abstract

Simple Summary Early identification of pregnancy enables farmers to separate pregnant and non-pregnant animals, adjust feeding, repeat breeding sooner, and reduce unnecessary handling. This review summarizes the development of early pregnancy diagnosis in livestock, from observation and palpation to hormone and pregnancy-associated protein tests, ultrasound, thermal imaging, optical sensing, wearable devices, and computer-assisted analysis. Conventional methods are inexpensive and practical but can be operator-dependent or reliable only after a certain stage of pregnancy. Laboratory tests and ultrasound generally provide higher accuracy, although they require sampling, equipment, or trained personnel. Emerging non-invasive sensors can collect information rapidly with less disturbance to animals, but their performance varies among species, farms, devices, hair conditions, and pregnancy stages. Combining complementary information, such as ultrasound, molecular markers, temperature, and optical signals, may improve reliability, but larger and independently validated studies are still needed. No single method is optimal for every livestock production system. Future tools should be accurate, affordable, easy to use on farms, and validated under diverse practical conditions.