A comparison of machine learning and logistic regression in modelling the association of body condition score and submission rate

Bates, Andrew; Saldias, Belen

doi:10.1016/j.prevetmed.2019.104765

Cited by 7 publications

(5 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We found no statistically significant associations between mean physiological and metabolic parameters measured between group of cows (according to embryo quality score) at each time point by ANOVA (Supporting Information: Table , FDR‐corrected p > 0.17). Other studies have reported a similar lack of predictive potential in parameters known to be involved in reproductive function (Bates & Saldias, 2019). The limited prognostic potential of these phenotypical variables underlines the necessity for higher resolution investigation, that is, molecular factors in ULF.…”

Section: Resultsmentioning

confidence: 74%

“…Table S1, FDR-corrected p > 0.17). Other studies have reported a similar lack of predictive potential in parameters known to be involved in reproductive function (Bates & Saldias, 2019). The limited prognostic potential of these phenotypical variables underlines the necessity for higher resolution investigation, that is, molecular factors in ULF.…”

Section: Physiological Changes In Early Postpartum Recoverymentioning

confidence: 78%

See 1 more Smart Citation

Metabolomic evolution of the postpartum dairy cow uterus

Aranciaga,

Ross,

Morton

et al. 2023

Molecular Reproduction Devel

View full text Add to dashboard Cite

High rates of early pregnancy loss are a critical issue in dairy herds, particularly in seasonal, grazing systems. Components of the uterine luminal fluid (ULF), on which the early embryo depends for sustenance and growth, partly determine early pregnancy losses. Here, changes in ULF from early to mid‐postpartum in crossbred dairy cows were explored, linking them with divergent embryo development. For this, the uteri of 87 cows at Day 7 of pregnancy at first and third estrus postpartum were flushed to collect ULF. Eighteen metabolites (chiefly organic acids and sugars) significantly varied in abundance across postpartum, indicating a molecular signature of physiological recovery consistent of the upregulation of pyrimidine metabolism and glycerophospholipid metabolism, and downregulation of pentose phosphate and taurine metabolism pathways. Joint pathway analysis of metabolomics data and a previously generated proteomics data set on the same ULF samples suggests key links between postpartum recovery and subsequent successful embryo development. These include upregulation of VEGFA and downregulation of metabolism, NRF2, T‐cell receptor, which appear to improve the ULF's capacity of sustaining normal embryo development, and a putative osmo‐protectant role of beta‐alanine. These relationships should be further investigated to develop tools to detect and reduce early pregnancy loss in dairy cows.

show abstract

Section: Resultsmentioning

confidence: 74%

Section: Physiological Changes In Early Postpartum Recoverymentioning

confidence: 78%

Metabolomic evolution of the postpartum dairy cow uterus

Aranciaga,

Ross,

Morton

et al. 2023

Molecular Reproduction Devel

View full text Add to dashboard Cite

show abstract

“…This work by Bates and Saldias [20] aimed to evaluate the accuracy with which several machine learning methods predicted the likelihood of service within three weeks after the expected onset of mating. They hypothesized that if the data contained complicated and unexplored connections or nonlinearity, many machine learning techniques would generate superior model performance than regression models.…”

Section: Literature Surveymentioning

confidence: 99%

Hybrid Deep Learning Approach for Enhanced Animal Breed Classification and Prediction

Khan,

Doohan,

Gupta

et al. 2023

View full text Add to dashboard Cite

The precise classification of animal breeds from image data is instrumental in real-time animal monitoring within forest ecosystems. Traditional computer vision methods have increasingly fallen short in accuracy due to the rapid progression of technology. To address these limitations, more advanced methodologies have emerged, significantly improving the accuracy of image classification, recognition, and segmentation tasks. The advent of "deep learning" has revolutionized various fields, particularly in object identification and recognition. Animal breed categorization is an important job in the field of image processing, and this research attempts to create a unique deep learning-based model for this purpose. The aim of this research is to devise efficient methodologies for image-based animal breed categorization to achieve superior accuracy levels. A hybrid deep learning model is proposed for animal breed prediction. The animal-10 dataset, obtained from Kaggle, serves as the empirical foundation for this study. The dataset underwent preprocessing, including edge deletion, normalization, and image scaling. Additionally, the animal images were converted into grayscale. Following this preprocessing phase, feature extraction was performed using two deep learning methods, namely VGG-19 and DenseNet121. The performance metrics, including accuracy, F1 score, recall, precision, and loss, were computed for the developed model using the Python simulation tool. Experimental results indicate that the proposed model outperforms existing current models in terms of these metrics. This research outcomes hold promising implications for the advancement of animal breed classification and prediction techniques.

show abstract

“…For instance, optimal BCS facilitates the mobilization of body reserves to support milk production ( Butler, 2014 ) and may greatly influence the reproductive success, e.g. increase the success rate for artificial insemination (AI) and conception ( Bates & Saldias, 2019 ). Cows with insufficient BCS and cows with a nadir of BCS in the post partum period below optimum are at risk to have lower conception rates ( Domecq et al., 1997 ; Pryce et al., 2001 ), to be more prone to metabolic (John R. Roche et al., 2013 ; Schuster et al., 2020 ) and infectious diseases ( Abunna et al., 2010 ; Jaja et al., 2017 ), e.g.…”

Section: Introductionmentioning

confidence: 99%