Improving Reproductive Rates in Egyptian Buffalo Using the OvSynch Protocol Under A deep Learning Prediction Model

This study aimed to enhance estrus and fertility rates in Behairy buffaloes using the GnRH-PGF2α-GnRH (GPG) protocol and introduce a deep learning prediction model under this protocol. Forty postpartum Egyptian buffaloes (55 days post-calving) were divided into two groups: G1 (control) and G2 (GPG protocol). The study found that the number of animals conceived increased significantly after the first service, from 20% to 70% in the control and GPG groups, respectively. Estrus was detected in 90% of GPG buffaloes, and conception and calving rates were higher in the GPG group. Male calf percentages also increased significantly in the GPG group compared to the control group. The GPG group showed an increase in calving birth weight but a significant decrease in calving interval and days open compared to the control group. The GPG group showed a substantial increase in zinc, magnesium, and calcium concentrations, while sodium concentration decreased. Estradiol 17-β significantly correlated with estrus response rate, while progesterone and sodium concentrations did not. The deep neural networks (DNN) demonstrated high accuracy in calf weight prediction 75%, with a tolerance of ± 8 Kg, and average daily milk yield accuracy of 80%, with a tolerance of ± 4.5 Kg. In conclusion, administering the GPG protocol may improve the production rate of Behairy Buffaloes under the deep learning prediction model.