Prediction of the likelihood of conception after the first or second insemination in Chinese Holstein cows using milk Fourier-transform infrared spectroscopy

• DOI: 10.3168/jds.2024-25269
• Publication Date: 2025-01-07T00:05:59Z
• Authors: Chu Chu; Peipei Wen; Weiqi Li; Yikai Fan; Zhuo Yang; Chao Du; Dongwei Wang; Liangkang Nan; Haitong Wang; Chunfang Li; Wenli Yu; Ahmed Sabek; Wan Wen; Guohua Hua; Junqing Ni; Yabin Ma; Shujun Zhang

<h2>ABSTRACT</h2> Accurate identification of cows' likelihood conception during the period from recent calving to first artificial insemination (AI) will provide assistance in managing fertility dairy cows and contribute economic prosperity sustainability farm. The purpose this study was use FTIR spectroscopy collected predict cow's AI, or second AI. This specifically focused on role spectral farm data at different time windows improving accuracy model for predicting From 2019 2023, information 10,873 Holstein China were collected, coupled with 21,928 data. First, classified into "good" "poor." Strategy 1 (S1) defined as conceived AI "poor" others. 2 (S2) Second, partial least squares discriminant analysis used establish models assessed using a cross-validation (CV) set herd-independent-external validation (HEV) set. also examining potential correlation between prediction collection by analyzing diagnostic performance 8 windows: '0 7 d postpartum (dpp)', '8 14 dpp', '15 21 '22 30 '31 45 '46 60 ' ≥ 61 before AI'. results showed that based S1 performed better while proximity AUC_CV AUC_HEV 0.621 0.633, respectively. Models S2 exhibited superior throughout late phase uterine involution. optimal developed dpp'. 0.644 0.660, respectively, which higher than S1. demonstrates ability conceive. within certain window accuracy, particularly 22 dpp 0 offers novel perspectives alternate approaches assessing cows, regularization farms, well precision management agriculture.