Predicting dry matter intake in Canadian Holstein dairy cattle using milk mid-infrared reflectance spectroscopy and other commonly available predictors via artificial neural networks
ABSTRACT: Dry matter intake (DMI) is a fundamental component of the animal's feed efficiency, but measuring DMI of individual cows is expensive. Mid-infrared reflectance spectroscopy (MIRS) on milk samples could be an inexpensive alternative to predict DMI. The objectives of this study were (1)...
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-10-01
|
| Series: | Journal of Dairy Science |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0022030222004817 |
| _version_ | 1798000783261696000 |
|---|---|
| author | Saeed Shadpour Tatiane C.S. Chud Dagnachew Hailemariam Hinayah R. Oliveira Graham Plastow Paul Stothard Jan Lassen Ransom Baldwin Filippo Miglior Christine F. Baes Dan Tulpan Flavio S. Schenkel |
| author_facet | Saeed Shadpour Tatiane C.S. Chud Dagnachew Hailemariam Hinayah R. Oliveira Graham Plastow Paul Stothard Jan Lassen Ransom Baldwin Filippo Miglior Christine F. Baes Dan Tulpan Flavio S. Schenkel |
| author_sort | Saeed Shadpour |
| collection | DOAJ |
| description | ABSTRACT: Dry matter intake (DMI) is a fundamental component of the animal's feed efficiency, but measuring DMI of individual cows is expensive. Mid-infrared reflectance spectroscopy (MIRS) on milk samples could be an inexpensive alternative to predict DMI. The objectives of this study were (1) to assess if milk MIRS data could improve DMI predictions of Canadian Holstein cows using artificial neural networks (ANN); (2) to investigate the ability of different ANN architectures to predict unobserved DMI; and (3) to validate the robustness of developed prediction models. A total of 7,398 milk samples from 509 dairy cows distributed over Canada, Denmark, and the United States were analyzed. Data from Denmark and the United States were used to increase the training data size and variability to improve the generalization of the prediction models over the lactation. For each milk spectra record, the corresponding weekly average DMI (kg/d), test-day milk yield (MY, kg/d), fat yield (FY, g/d), and protein yield (PY, g/d), metabolic body weight (MBW), age at calving, year of calving, season of calving, days in milk, lactation number, country, and herd were available. The weekly average DMI was predicted with various ANN architectures using 7 predictor sets, which were created by different combinations MY, FY, PY, MBW, and MIRS data. All predictor sets also included age of calving and days in milk. In addition, the classification effects of season of calving, country, and lactation number were included in all models. The explored ANN architectures consisted of 3 training algorithms (Bayesian regularization, Levenberg-Marquardt, and scaled conjugate gradient), 2 types of activation functions (hyperbolic tangent and linear), and from 1 to 10 neurons in hidden layers). In addition, partial least squares regression was also applied to predict the DMI. Models were compared using cross-validation based on leaving out 10% of records (validation A) and leaving out 10% of cows (validation B). Superior fitting statistics of models comprising MIRS information compared with the models fitting milk, fat and protein yields suggest that other unknown milk components may help explain variation in weekly average DMI. For instance, using MY, FY, PY, and MBW as predictor variables produced a predictive accuracy (r) ranging from 0.510 to 0.652 across ANN models and validation sets. Using MIRS together with MY, FY, PY, and MBW as predictors resulted in improved fitting (r = 0.679–0.777). Including MIRS data improved the weekly average DMI prediction of Canadian Holstein cows, but it seems that MIRS predicts DMI mostly through its association with milk production traits and its utility to estimate a measure of feed efficiency that accounts for the level of production, such as residual feed intake, might be limited and needs further investigation. The better predictive ability of nonlinear ANN compared with linear ANN and partial least squares regression indicated possible nonlinear relationships between weekly average DMI and the predictor variables. In general, ANN using Bayesian regularization and scaled conjugate gradient training algorithms yielded slightly better weekly average DMI predictions compared with ANN using the Levenberg-Marquardt training algorithm. |
| first_indexed | 2024-04-11T11:25:50Z |
| format | Article |
| id | doaj.art-4f155cd589034c69a87caba4294123c6 |
| institution | Directory Open Access Journal |
| issn | 0022-0302 |
| language | English |
| last_indexed | 2024-04-11T11:25:50Z |
| publishDate | 2022-10-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of Dairy Science |
| spelling | doaj.art-4f155cd589034c69a87caba4294123c62022-12-22T04:26:17ZengElsevierJournal of Dairy Science0022-03022022-10-011051082578271Predicting dry matter intake in Canadian Holstein dairy cattle using milk mid-infrared reflectance spectroscopy and other commonly available predictors via artificial neural networksSaeed Shadpour0Tatiane C.S. Chud1Dagnachew Hailemariam2Hinayah R. Oliveira3Graham Plastow4Paul Stothard5Jan Lassen6Ransom Baldwin7Filippo Miglior8Christine F. Baes9Dan Tulpan10Flavio S. Schenkel11Centre for Genetic Improvement of Livestock, Department of Animal Biosciences, University of Guelph, Guelph, Ontario, N1G 2W1, CanadaCentre for Genetic Improvement of Livestock, Department of Animal Biosciences, University of Guelph, Guelph, Ontario, N1G 2W1, CanadaDepartment of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Alberta, T6G 2P5, CanadaCentre for Genetic Improvement of Livestock, Department of Animal Biosciences, University of Guelph, Guelph, Ontario, N1G 2W1, CanadaDepartment of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Alberta, T6G 2P5, CanadaDepartment of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Alberta, T6G 2P5, CanadaCenter for Quantitative Genetics and Genomics, Aarhus University, Tjele, 8830, DenmarkAnimal Genomics and Improvement Laboratory, Beltsville Agricultural Research Center, Agricultural Research Service, US Department of Agriculture, Beltsville, MD 20705Centre for Genetic Improvement of Livestock, Department of Animal Biosciences, University of Guelph, Guelph, Ontario, N1G 2W1, Canada; Lactanet Canada, Guelph, Ontario, N1K 1E5, CanadaCentre for Genetic Improvement of Livestock, Department of Animal Biosciences, University of Guelph, Guelph, Ontario, N1G 2W1, Canada; Institute of Genetics, Vetsuisse Faculty, University of Bern, Bern, Switzerland 3001Centre for Genetic Improvement of Livestock, Department of Animal Biosciences, University of Guelph, Guelph, Ontario, N1G 2W1, CanadaCentre for Genetic Improvement of Livestock, Department of Animal Biosciences, University of Guelph, Guelph, Ontario, N1G 2W1, Canada; Corresponding authorABSTRACT: Dry matter intake (DMI) is a fundamental component of the animal's feed efficiency, but measuring DMI of individual cows is expensive. Mid-infrared reflectance spectroscopy (MIRS) on milk samples could be an inexpensive alternative to predict DMI. The objectives of this study were (1) to assess if milk MIRS data could improve DMI predictions of Canadian Holstein cows using artificial neural networks (ANN); (2) to investigate the ability of different ANN architectures to predict unobserved DMI; and (3) to validate the robustness of developed prediction models. A total of 7,398 milk samples from 509 dairy cows distributed over Canada, Denmark, and the United States were analyzed. Data from Denmark and the United States were used to increase the training data size and variability to improve the generalization of the prediction models over the lactation. For each milk spectra record, the corresponding weekly average DMI (kg/d), test-day milk yield (MY, kg/d), fat yield (FY, g/d), and protein yield (PY, g/d), metabolic body weight (MBW), age at calving, year of calving, season of calving, days in milk, lactation number, country, and herd were available. The weekly average DMI was predicted with various ANN architectures using 7 predictor sets, which were created by different combinations MY, FY, PY, MBW, and MIRS data. All predictor sets also included age of calving and days in milk. In addition, the classification effects of season of calving, country, and lactation number were included in all models. The explored ANN architectures consisted of 3 training algorithms (Bayesian regularization, Levenberg-Marquardt, and scaled conjugate gradient), 2 types of activation functions (hyperbolic tangent and linear), and from 1 to 10 neurons in hidden layers). In addition, partial least squares regression was also applied to predict the DMI. Models were compared using cross-validation based on leaving out 10% of records (validation A) and leaving out 10% of cows (validation B). Superior fitting statistics of models comprising MIRS information compared with the models fitting milk, fat and protein yields suggest that other unknown milk components may help explain variation in weekly average DMI. For instance, using MY, FY, PY, and MBW as predictor variables produced a predictive accuracy (r) ranging from 0.510 to 0.652 across ANN models and validation sets. Using MIRS together with MY, FY, PY, and MBW as predictors resulted in improved fitting (r = 0.679–0.777). Including MIRS data improved the weekly average DMI prediction of Canadian Holstein cows, but it seems that MIRS predicts DMI mostly through its association with milk production traits and its utility to estimate a measure of feed efficiency that accounts for the level of production, such as residual feed intake, might be limited and needs further investigation. The better predictive ability of nonlinear ANN compared with linear ANN and partial least squares regression indicated possible nonlinear relationships between weekly average DMI and the predictor variables. In general, ANN using Bayesian regularization and scaled conjugate gradient training algorithms yielded slightly better weekly average DMI predictions compared with ANN using the Levenberg-Marquardt training algorithm.http://www.sciencedirect.com/science/article/pii/S0022030222004817dry matter intakemachine learningmid-infrared reflectance spectroscopymilk |
| spellingShingle | Saeed Shadpour Tatiane C.S. Chud Dagnachew Hailemariam Hinayah R. Oliveira Graham Plastow Paul Stothard Jan Lassen Ransom Baldwin Filippo Miglior Christine F. Baes Dan Tulpan Flavio S. Schenkel Predicting dry matter intake in Canadian Holstein dairy cattle using milk mid-infrared reflectance spectroscopy and other commonly available predictors via artificial neural networks Journal of Dairy Science dry matter intake machine learning mid-infrared reflectance spectroscopy milk |
| title | Predicting dry matter intake in Canadian Holstein dairy cattle using milk mid-infrared reflectance spectroscopy and other commonly available predictors via artificial neural networks |
| title_full | Predicting dry matter intake in Canadian Holstein dairy cattle using milk mid-infrared reflectance spectroscopy and other commonly available predictors via artificial neural networks |
| title_fullStr | Predicting dry matter intake in Canadian Holstein dairy cattle using milk mid-infrared reflectance spectroscopy and other commonly available predictors via artificial neural networks |
| title_full_unstemmed | Predicting dry matter intake in Canadian Holstein dairy cattle using milk mid-infrared reflectance spectroscopy and other commonly available predictors via artificial neural networks |
| title_short | Predicting dry matter intake in Canadian Holstein dairy cattle using milk mid-infrared reflectance spectroscopy and other commonly available predictors via artificial neural networks |
| title_sort | predicting dry matter intake in canadian holstein dairy cattle using milk mid infrared reflectance spectroscopy and other commonly available predictors via artificial neural networks |
| topic | dry matter intake machine learning mid-infrared reflectance spectroscopy milk |
| url | http://www.sciencedirect.com/science/article/pii/S0022030222004817 |
| work_keys_str_mv | AT saeedshadpour predictingdrymatterintakeincanadianholsteindairycattleusingmilkmidinfraredreflectancespectroscopyandothercommonlyavailablepredictorsviaartificialneuralnetworks AT tatianecschud predictingdrymatterintakeincanadianholsteindairycattleusingmilkmidinfraredreflectancespectroscopyandothercommonlyavailablepredictorsviaartificialneuralnetworks AT dagnachewhailemariam predictingdrymatterintakeincanadianholsteindairycattleusingmilkmidinfraredreflectancespectroscopyandothercommonlyavailablepredictorsviaartificialneuralnetworks AT hinayahroliveira predictingdrymatterintakeincanadianholsteindairycattleusingmilkmidinfraredreflectancespectroscopyandothercommonlyavailablepredictorsviaartificialneuralnetworks AT grahamplastow predictingdrymatterintakeincanadianholsteindairycattleusingmilkmidinfraredreflectancespectroscopyandothercommonlyavailablepredictorsviaartificialneuralnetworks AT paulstothard predictingdrymatterintakeincanadianholsteindairycattleusingmilkmidinfraredreflectancespectroscopyandothercommonlyavailablepredictorsviaartificialneuralnetworks AT janlassen predictingdrymatterintakeincanadianholsteindairycattleusingmilkmidinfraredreflectancespectroscopyandothercommonlyavailablepredictorsviaartificialneuralnetworks AT ransombaldwin predictingdrymatterintakeincanadianholsteindairycattleusingmilkmidinfraredreflectancespectroscopyandothercommonlyavailablepredictorsviaartificialneuralnetworks AT filippomiglior predictingdrymatterintakeincanadianholsteindairycattleusingmilkmidinfraredreflectancespectroscopyandothercommonlyavailablepredictorsviaartificialneuralnetworks AT christinefbaes predictingdrymatterintakeincanadianholsteindairycattleusingmilkmidinfraredreflectancespectroscopyandothercommonlyavailablepredictorsviaartificialneuralnetworks AT dantulpan predictingdrymatterintakeincanadianholsteindairycattleusingmilkmidinfraredreflectancespectroscopyandothercommonlyavailablepredictorsviaartificialneuralnetworks AT flaviosschenkel predictingdrymatterintakeincanadianholsteindairycattleusingmilkmidinfraredreflectancespectroscopyandothercommonlyavailablepredictorsviaartificialneuralnetworks |