Validation of an Automated Body Condition Scoring Camera for Dairy Cattle
Grade Level at Time of Presentation
Junior
Major
Animal Science
Minor
None
Institution
University of Kentucky
KY House District #
80
KY Senate District #
25
Faculty Advisor/ Mentor
Jeffrey Bewley, PhD; Ms. Carissa Truman
Department
Department of Animal and Food Sciences
Abstract
Validation of an Automated Body Condition Scoring Camera for Dairy Cattle
I.L. Mullins, C.M. Truman, and J.M. Bewley, Animal and Foods Sciences, University of Kentucky, Lexington, KY
Body condition scoring (BCS) is a visual estimate of subcutaneous fat reserves. The scoring system ranges from 1 to 5, in 0.25 increments, with 1 being extremely thin and 5 being morbidly obese. The desirable range for lactating dairy cows varies, dependent on the time in lactation and should be monitored at multiple time points for the greatest managerial impact. Undesirable body condition scores have been shown to negatively impact milk production, disease, and reproduction. Manually scoring BCS has proven beneficial yet requires time for data collection, entry, and analysis. Recently, a camera that automatically scores and records body condition scores of cattle has been released to the market (DeLaval International AB,Tumba, Sweden). The objective of this study was to validate the correlation of the cameras’ automated scores with conventional manual scores. Three experienced individuals scored the cows manually, which were recorded and averaged by cow, to compare with the automated scores. All data analysis was performed using SAS 9.3 (SAS Institute Inc., Cary, NC). Using PROC UNIVARIATE, descriptive statistics were found for the averaged manual and automated scores. The average score for all the manual scores was 3.38 ± 0.48 (mean ± SD) The average automated score was 3.27 ± 0.27 (mean ± SD). The scores were compared using PROC CORR and the correlation coefficient was 0.78 (P<0.001). The correlation demonstrated a strong positive relationship between the manually obtained scores and the automated scores. The agreement between the scores indicates that the automated system can provide BCS that corresponds with a conventional manual scoring system. The automated system may encourage more producers to adopt BCS into their management practices. Using BCS, producers will be able to make management decisions to improve milk yield, reproduction success, and lower disease occurrence.
Validation of an Automated Body Condition Scoring Camera for Dairy Cattle
Validation of an Automated Body Condition Scoring Camera for Dairy Cattle
I.L. Mullins, C.M. Truman, and J.M. Bewley, Animal and Foods Sciences, University of Kentucky, Lexington, KY
Body condition scoring (BCS) is a visual estimate of subcutaneous fat reserves. The scoring system ranges from 1 to 5, in 0.25 increments, with 1 being extremely thin and 5 being morbidly obese. The desirable range for lactating dairy cows varies, dependent on the time in lactation and should be monitored at multiple time points for the greatest managerial impact. Undesirable body condition scores have been shown to negatively impact milk production, disease, and reproduction. Manually scoring BCS has proven beneficial yet requires time for data collection, entry, and analysis. Recently, a camera that automatically scores and records body condition scores of cattle has been released to the market (DeLaval International AB,Tumba, Sweden). The objective of this study was to validate the correlation of the cameras’ automated scores with conventional manual scores. Three experienced individuals scored the cows manually, which were recorded and averaged by cow, to compare with the automated scores. All data analysis was performed using SAS 9.3 (SAS Institute Inc., Cary, NC). Using PROC UNIVARIATE, descriptive statistics were found for the averaged manual and automated scores. The average score for all the manual scores was 3.38 ± 0.48 (mean ± SD) The average automated score was 3.27 ± 0.27 (mean ± SD). The scores were compared using PROC CORR and the correlation coefficient was 0.78 (P<0.001). The correlation demonstrated a strong positive relationship between the manually obtained scores and the automated scores. The agreement between the scores indicates that the automated system can provide BCS that corresponds with a conventional manual scoring system. The automated system may encourage more producers to adopt BCS into their management practices. Using BCS, producers will be able to make management decisions to improve milk yield, reproduction success, and lower disease occurrence.