Determination of ewe behaviour around lambing time and prediction of parturition 7 days prior to lambing by tri-axial accelerometer sensors in an extensive farming system

Context Lamb loss and dyctocia are two major challenges in extensive farming systems. While visual observation can be impractical due to the large sizes of paddocks, number animals high labour cost, wearable sensors used monitor behaviour ewes as there might changes their activities prior lambing. This provides sufficient time for farm manager nurse those that at risk dyctocia. Aim The objective this study was determine whether a pregnant ewe could predict parturition. Methods Two separate trials were conducted: first trial (T1), with 32 ewes, included human/video observations, second (T2), 165 conducted no humans present, emulate real settings. fitted tri-axial accelerometer by means halters. Three-dimensional movement data collected period least 7 14 days T1 T2 respectively. sensor units retrieved, downloaded using ActiGraph software. Ewe determined through support vector machine learning (SVM) algorithm, including licking, grazing, rumination, walking, idling. behaviours predicted analysis compared (video recordings), synchronisation validate results. Deep neural-network algorithms lambing time. Key results concordance percentages between 90 ± 11, 81 15, 95 10, 96 6, 93 8% s.d. idling, walking deep-learning model 90% confidence via quantile regression method, which interpereted prediction intervals, shows reasonable approximately 240 h before actual events. Conclusion It possible parturition up 10 Implications around has direct effect on survival lambs therefore plays an important part animal management. knowledge improve productivity sheep considerably decrease lamb mortality rates.