Effect of estrous detection strategy on pregnancy outcomes of lactating Holstein cows receiving artificial insemination and embryo transfer

Machine Learning to Predict Pregnancy in Dairy Cows: An Approach Integrating Automated Activity Monitoring and On-Farm Data

Automated activity monitoring (AAM) systems are critical in the dairy industry for detecting estrus and optimizing the timing of artificial insemination (AI), thus enhancing pregnancy success rates in cows. This study developed a predictive model to improve pregnancy success by integrating AAM data with cow-specific and environmental factors. Utilizing data from 1,054 cows, this study compared the pregnancy outcomes between two AI timings-8 or 10 h post-AAM alarm. Variables such as age, parity, body condition, locomotion, and vaginal discharge scores, peripartum diseases, the breeding program, the bull used for AI, milk production at the time of AI, and environmental conditions (season, relative humidity, and temperature-humidity index) were considered alongside the AAM data on rumination, activity, and estrus intensity. Six predictive models were assessed to determine their efficacy in predicting pregnancy success: logistic regression, Bagged AdaBoost algorithm, linear discriminant, random forest, support vector machine, and Bagged Classification Tree. Integrating the on-farm data with AAM significantly enhanced the pregnancy prediction accuracy at AI compared to using AAM data alone. The random forest models showed a superior performance, with the highest Kappa statistic and lowest false positive rates. The linear discriminant and logistic regression models demonstrated the best accuracy, minimal false negatives, and the highest area under the curve. These findings suggest that combining on-farm and AAM data can significantly improve reproductive management in the dairy industry.

Reproduction in the era of genomics and automation

Much progress has been made in the reproductive efficiency of lactating dairy cows across the USA in the past 20years. The standardisation of evaluation of reproductive efficiency, particularly with greater focus on metrics with lesser momentum and less lag-time such as 21-day pregnancy rates (21-day PR), and the recognition that subpar reproductive efficiency negatively impacted profitability were major drivers for the changes that resulted in such progress. Once it became evident that the genetic selection of cattle for milk yield regardless of fertility traits was associated with reduced fertility, geneticists raced to identify fertility traits that could be incorporated in genetic selection programs with the hopes of improving fertility of lactating cows. Concurrently, reproductive physiologists developed ovulation synchronisation protocols such that after sequential treatment with exogenous hormones, cows could be inseminated at fixed time and without detection of oestrus and still achieve acceptable pregnancy per service. These genetic and reproductive management innovations, concurrently with improved husbandry and nutrition of periparturient cows, quickly started to move reproductive efficiency of lactating dairy cows to an upward trend that continues today. Automation has been adopted in Israel and European countries for decades, but only recently have these automated systems been more widely adopted in the USA. The selection of dairy cattle based on genetic indexes that result in positive fertility traits (e.g. daughter pregnancy rate) is positively associated with follicular growth, resumption of ovarian cycles postpartum, body condition score and insulin-like growth factor 1 concentration postpartum, and intensity of oestrus. Collectively, these positive physiological characteristics result in improved reproductive performance. Through the use of automated monitoring devices (AMD), it is possible to identify cows that resume cyclicity sooner after calving and have more intense oestrus postpartum, which are generally cows that have a more successful periparturient period. Recent experiments have demonstrated that it may be possible to adopt targeted reproductive management, utilising ovulation synchronisation protocols for cows that do not have intense oestrus postpartum and relying more heavily on insemination at AMD-detected oestrus for cows that display an intense oestrus postpartum. This strategy is likely to result in tailored hormonal therapy that will be better accepted by the public, will increase the reliance on oestrus for insemination, will improve comfort and reduce labour by reducing the number of injections cows receive in a lactation, and will allow for faster decisions about cows that should not be eligible for insemination.

Targeted reproductive management for lactating Holstein cows: Reducing the reliance on exogenous reproductive hormones

Adoption of automated monitoring devices (AMD) affords the opportunity to tailor reproductive management according to the cow's needs. We hypothesized that a targeted reproductive management (TRM) would reduce the use of reproductive hormones while increasing the percentage of cows pregnant 305 d in milk (DIM). Holstein cows from 2 herds (n = 1,930) were fitted with an AMD at 251.0 ± 0.4 d of gestation. Early-postpartum estrus characteristics (EPEC; intense estrus = heat index ≥70; 0 = minimum, 100 = maximum) of multiparous cows were evaluated at 40 (herd 1) or 41 (herd 2) DIM and EPEC of primiparous cows were evaluated at 54 (herd 1) or 55 (herd 2) DIM. Control cows received the first artificial insemination at fixed time (TAI; primiparous, herd 1 = 82 and herd 2 = 83 DIM; multiparous, herd 1 = 68 and herd 2 = 69 DIM) following the Double-Ovsynch (DOV) protocol. Cows enrolled in the TRM treatment were managed as follows: (1) cows with at least one intense estrus were inseminated upon AMD detected estrus for 42 d and, if not inseminated, were enrolled in the DOV protocol; and (2) cows without an intense estrus were enrolled in the DOV protocol at the same time as cows in the control treatment. Control cows were re-inseminated based on visual or patch aided detection of estrus, whereas TRM cows were re-inseminated as described for control cows with the aid of the AMD. Cows received a GnRH injection 27 ± 3 d after insemination and, if diagnosed as nonpregnant, completed the 5-d Cosynch protocol and received TAI 35 ± 3 d after insemination. Among cows in the TRM treatment, 55.8 and 42.9% of primiparous and multiparous cows, respectively, received the first insemination in spontaneous estrus. The interaction between treatment and parity affected pregnancy 67 d after the first AI (primiparous: control = 37.6%, TRM = 27.4%; multiparous: control = 41.0%, TRM = 44.7%). The TRM treatment increased re-insemination in estrus (control = 48.3%, TRM = 70.5%). Pregnancy 67 d after re-inseminations tended to be affected by the interaction between treatment and EPEC (no intense estrus: control = 25.3%, TRM = 32.0%; intense estrus: control = 32.9%, TRM = 32.2%). The interaction between treatment and EPEC affected pregnancy by 305 DIM (no intense estrus: control = 80.8%, TRM = 88.2%; intense estrus: control = 87.1%, TRM = 86.1%). Treatment did not affect the number of reproductive hormone treatments among cows that had not had an intense estrus (control = 10.5 ± 0.3, TRM = 9.1 ± 0.2 treatments/cow), but cows in the TRM treatment that had an intense estrus received fewer reproductive hormone treatments than cows in the control treatment (2.0 ± 0.1 vs. 9.6 ± 0.2 treatments/cow). Selecting multiparous cows for first AI in estrus based on EPEC reduced the use of reproductive hormones without impairing the likelihood of pregnancy to first AI. The use of AMD for re-insemination expedited the establishment of pregnancy among cows that did not display an intense estrus early postpartum.

Review: Some challenges and unrealized opportunities toward widespread use of the in vitro-produced embryo in cattle production

The embryo produced by in vitro oocyte maturation, fertilization, and embryonic development is an important resource for genetic improvement and has the potential to improve female fertility and to be programmed to produce offspring with superior ability for health and production. The cultured embryo is also an important component of several realized and potential technologies such as gene editing, somatic cell nuclear cloning, stem cell technologies and gamete generation in vitro. Full realization of the opportunities afforded by the in vitro-produced embryo will require overcoming some technical obstacles to cost-effective implementation of an embryo transfer program. Among the research goals for improving the penetration of embryo transfer in the cattle industry are development of methods to increase the supply of oocytes from genetically elite females, enhance the proportion of oocytes that become transferrable embryos, improve the fraction of embryos that establish pregnancy after transfer, reduce pregnancy wastage after pregnancy diagnosis, and identify culture conditions to optimize postnatal phenotype.

A Recognition Method of Ewe Estrus Crawling Behavior Based on Multi-Target Detection Layer Neural Network

There are some problems with estrus detection in ewes in large-scale meat sheep farming: mainly, the manual detection method is labor-intensive and the contact sensor detection method causes stress reactions in ewes. To solve the abovementioned problems, we proposed a multi-objective detection layer neural network-based method for ewe estrus crawling behavior recognition. The approach we proposed has four main parts. Firstly, to address the problem of mismatch between our constructed ewe estrus dataset and the YOLO v3 anchor box size, we propose to obtain a new anchor box size by clustering the ewe estrus dataset using the K-means++ algorithm. Secondly, to address the problem of low model recognition precision caused by small imaging of distant ewes in the dataset, we added a 104 × 104 target detection layer, making the total target detection layer reach four layers, strengthening the model's ability to learn shallow information and improving the model's ability to detect small targets. Then, we added residual units to the residual structure of the model, so that the deep feature information of the model is not easily lost and further fused with the shallow feature information to speed up the training of the model. Finally, we maintain the aspect ratio of the images in the data-loading module of the model to reduce the distortion of the image information and increase the precision of the model. The experimental results show that our proposed model has 98.56% recognition precision, while recall was 98.04%, F1 value was 98%, mAP was 99.78%, FPS was 41 f/s, and model size was 276 M, which can meet the accurate and real-time recognition of ewe estrus behavior in large-scale meat sheep farming.

Effect of GnRH administered at the time of artificial insemination for cows detected in estrus by conventional estrus detection or an automated activity-monitoring system

The objectives of this study were to determine the effects of GnRH at the time of artificial insemination (AI) on ovulation, progesterone 7 d post-AI, and pregnancy in cows detected in estrus using traditional methods (tail chalk removal and mount acceptance visualization) or an automated activity-monitoring (AAM) system. We hypothesized that administration of GnRH at the time of AI would increase ovulation rate, plasma progesterone post-AI, and pregnancy per AI (P/AI) in cows detected in estrus. In experiment 1, Holstein cows (n = 398) were blocked by parity and randomly assigned to receive an injection of GnRH at the time of estrus detection/AI (GnRH, n = 197) or to remain untreated (control, n = 201) on 4 farms. The GnRH was administered as 100 µg of gonadorelin acetate. Ovarian structures and plasma progesterone were assessed in a subset of cows (GnRH, n = 52; control, n = 55) in experiment 1 at the time of AI and 7 d later. In experiment 2, a group of 409 cows in an AAM farm were enrolled as described for experiment 1 (GnRH, n = 207; control, n = 202). Data were categorized for parity (primiparous vs. multiparous), season (cool vs. warm), number of services (first vs. > first), DIM (>150 DIM vs. ≤150 DIM), and for AAM cows in experiment 2 for activity level (high: 90-100 index vs. low: 35-89 index). Pregnancy diagnosis was performed between 32 and 45 d post-AI (P1) and 60 to 115 d post-AI (P2). In experiment 1, there was no difference in plasma progesterone at day of estrus detection (control = 0.09 ng/mL vs. GnRH = 0.16 ng/mL), 7 d later (control = 2.03 ng/mL vs. GnRH = 2.18 ng/mL), and ovulation rate (GnRH = 83.2% vs. control = 77.9%) between treatments. There were no effects of GnRH in experiment 1 for P/AI at P1 (control = 43.3% vs. GnRH = 38.6%), P2 (control = 38.4% vs. GnRH = 34.5%), and for pregnancy loss (control = 9.8% vs. GnRH = 8.2%). In experiment 2, there were no effects of GnRH for P/AI at P1 (control = 39.6% vs. GnRH = 40.1%), P2 (control = 35.0% vs. GnRH = 37.4%), and for pregnancy loss (control = 9.5% vs. GnRH = 6.2%). There was a tendency for a parity effect on P/AI for P1, but not P2 or for pregnancy loss. High-activity cows had greater P/AI in P1 (low activity = 27.9% vs. high activity = 44.1%), P2 (low activity = 21.8% vs. high activity = 41.2%), and lower pregnancy loss (low activity = 20.7% vs. high activity = 5.1%), but there were no interactions between treatment and activity level. The current study did not support the use of GnRH at estrus detection to improve ovulatory response, progesterone 1 wk post-AI, and P/AI. More research is needed to investigate the relationship between GnRH at the time of AI and activity level in herds using AAM systems.

Genomic Evaluation of Primiparous High-Producing Dairy Cows: Inbreeding Effects on Genotypic and Phenotypic Production-Reproductive Traits

Simple Summary

Improving the genomic prediction methodologies in high-producing dairy cattle is a key factor for the selection of suitable individuals to ensure better productivity. However, the most advanced prediction tools based on genotyping show ~75% reliability. Nowadays, the incorporation of new indices to genomic prediction methods, such as the Inbreeding Index (II), can significantly facilitate the selection of reliable production and reproductive traits for progeny selection. Thus, the objective of this study was to determine the impact of II (low: LI and high: HI), based on genomic analysis, and its effect on production and reproductive phenotypic traits in high-producing primiparous dairy cows. Individuals with II between ≥2.5 and ≤5.0 have shown up to a two-fold increase in negative correlations comparing LI versus HI genomic production and reproductive parameters, severely affecting important traits such as Milk Production at 305 d, Protein Production at 305 d, Fertility Index, and Daughter Pregnancy Rate. Therefore, high-producing dairy cows face an increased risk of negative II-derived effects in their selection programs, particularly at II ≥ 2.5.

Abstract

The main objective of this study was to analyze the effects of the inbreeding degree in high-producing primiparous dairy cows genotypically and phenotypically evaluated and its impacts on production and reproductive parameters. Eighty Holstein–Friesian primiparous cows (age: ~26 months; ~450 kg body weight) were previously genomically analyzed to determine the Inbreeding Index (II) and were divided into two groups: low inbreeding group (LI: <2.5; n = 40) and high inbreeding group (HI: ≥2.5 and ≤5.0; n = 40). Genomic determinations of production and reproductive parameters (14 in total), together with analyses of production (12) and reproductive (11) phenotypic parameters (23 in total) were carried out. Statistically significant differences were obtained between groups concerning the genomic parameters of Milk Production at 305 d and Protein Production at 305 d and the reproductive parameter Daughter Calving Ease, the first two being higher in cows of the HI group and the third lower in the LI group (p < 0.05). For the production phenotypic parameters, statistically significant differences were observed between both groups in the Total Fat, Total Protein, and Urea parameters, the first two being higher in the LI group (p < 0.05). Also, significant differences were observed in several reproductive phenotypic parameters, such as Number of Services per Conception, Calving to Conception Interval, Days Open Post Service, and Current Inter-Partum Period, all of which negatively influenced the HI group (p < 0.05). In addition, correlation analyses were performed between production and reproductive genomic parameters separately and in each consanguinity group. The results showed multiple positive and negative correlations between the production and reproductive parameters independently of the group analyzed, being these correlations more remarkable for the reproductive parameters in the LI group and the production parameters in the HI group (p < 0.05). In conclusion, the degree of inbreeding significantly influenced the results, affecting different genomic and phenotypic production and reproductive parameters in high-producing primiparous cows. The determination of the II in first-calf heifers is crucial to evaluate the negative effects associated with homozygosity avoiding an increase in inbreeding depression on production and reproductive traits.