Machine learning algorithms for the prediction of conception success to a given insemination in lactating dairy cows

Abstract

The ability to accurately predict the conception outcome for a future mating would be of considerable benefit for producers in deciding what mating plan (i.e., expensive semen or less expensive semen) to implement for a given cow. The objective of the present study was to use herd- and cow-level factors to predict the likelihood of conception success to a given insemination (i.e., conception outcome not including embryo loss); of particular interest in the present study was the usefulness of milk mid-infrared (MIR) spectral data in augmenting the accuracy of the prediction model. A total of 4,341 insemination records with conception outcome information from 2,874 lactations on 1,789 cows from 7 research herds for the years 2009 to 2014 were available. The data set was separated into a calibration data set and a validation data set using either of 2 approaches: (1) the calibration data set contained records from all 7 farms for the years 2009 to 2011, inclusive, and the validation data set included data from the 7 farms for the years 2012 to 2014, inclusive, or (2) the calibration data set contained records from 5 farms for all 6 yr and the validation data set contained information from the other 2 farms for all 6 yr. The prediction models were developed with 8 different machine learning algorithms in the calibration data set using standard 10-times 10-fold cross-validation and also by evaluating in the validation data set. The area under curve (AUC) for the receiver operating curve varied from 0.487 to 0.675 across the different algorithms and scenarios investigated. Logistic regression was generally the best-performing algorithm. The AUC was generally inferior for the external validation data sets compared with the calibration data sets. The inclusion of milk MIR in the prediction model generally did not improve the accuracy of prediction. Despite the fair AUC for predicting conception outcome under the different scenarios investigated, the model provided a reasonable prediction of the likelihood of conception success when the high predicted probability instances were considered; a conception rate of 85% was evident in the top 10% of inseminations ranked on predicted probability of conception success in the validation data set.

Associations between postpartum phenotypes, cow factors, genetic traits, and reproductive performance in seasonal-calving, pasture-based lactating dairy cows

“…Recent studies from our group examining the genetic merit for fertility traits have reported greater DMI and BCS, superior metabolic status and uterine health (Cummins et al, 2012a;Moore et al, 2014a;Moran et al, 2017), earlier resumption of ovarian cyclicity, and greater luteal phase circulating concentrations of progesterone (Cummins et al, 2012b;Moore et al, 2014b) in cows with good genetic merit for fertility traits compared with cows with poor genetic merit for fertility traits. The results arising from this study highlight the strong associations between fertility subindex and PTA for both calving interval and survival with reproductive performance, in agreement with Fenlon et al (2017) and Hempstalk et al (2015). This highlights the importance of selecting for fertility traits and long-term gains that can be achieved in herd phenotypic reproductive performance.…”

Classifying the fertility of dairy cows using milk mid-infrared spectroscopy

“…For example, sexed or premium bull semen could be used for cows predicted to have a high likelihood of conception, whereas cows with predicted poor fertility could be mated using semen from beef bulls, multiple doses, or semen from bulls of known high genetic merit for fertility. To our knowledge, Grzesiak et al (2010), Shahinfar et al (2014), and Hempstalk et al (2015) are the only authors who have reported the ability of some on-farm data to predict likelihood of conception to a given insemination of dairy cows. The value of the prediction accuracy ranged between 0.66 (Hempstalk et al, 2015) and 0.91 (Grzesiak et al, 2010).…”

“…To our knowledge, Grzesiak et al (2010), Shahinfar et al (2014), and Hempstalk et al (2015) are the only authors who have reported the ability of some on-farm data to predict likelihood of conception to a given insemination of dairy cows. The value of the prediction accuracy ranged between 0.66 (Hempstalk et al, 2015) and 0.91 (Grzesiak et al, 2010). Unfortunately, some of the important variables used in these studies might be difficult to obtain on-farm (e.g., BCS and BW) or cannot be predicted a priori (e.g., year).…”

Standardization of milk mid-infrared spectrometers for the transfer and use of multiple models

“…Other studies have focused on FT-MIR spectra to build equations predicting technological properties of milk such as milk acidity ), ability to coagulate, firmness of curd, or cheese yield (Dal Zotto et al, 2008;Colinet et al, 2015). Recent work has directly considered the FT-MIR spectrum of milk as a reflection of cows' status, with FT-MIR equations being developed to predict methane emissions of dairy cows (Dehareng et al, 2012;Vanlierde et al, 2016), likelihood of conception (Hempstalk et al, 2015), body energy status , energy intake and efficiency (McParland et al, 2014). In the work of Lainé et al (2017), the spectrum is even considered as a response for which the effect of pregnancy is evaluated.…”