LACTOPEROXIDASE POTENTIAL IN THE DIAGNOSIS OF SUBCLINICAL MASTITIS IN BOVINE VIA IMAGE PROCESSING
udder health, immune system, RGB images, principal component analysis
Due to the economic damages to dairy cattle in the order of 200 dollars for each cow per year, subclinical mastitis has increasingly been the target of research aimed at the control and early diagnosis, since in addition to affecting milk production and composition, modifies the enzymatic concentration, causing losses in the production of derivatives and reduction of the shelf life. Lactoperoxidase is one of the enzymes that has its activity altered when the animal is affected with mastitis, so, studies show that there is a positive correlation between lactoperoxidase and somatic cell count (CCS). The objective of this work was to measure the lactoperoxidase quantitatively and qualitatively and, through image analysis, to verify the potentiality of the enzyme in the detection of subclinical mastitis in cattle. Fifty samples of milk (34 from Jersey, 08 from Gir and 08 from Girolando cows) were analyzed for CCS and enzymatic activity. The sample evaluation system consisted in correlating the color presented by the sample with the enzymatic concentration using the RGB (Red - Green - Blue) color system. Data analysis was performed using the statistical platform R 3.5.0. The PCA model (Packet Chemometrics With R6 and factoextra7) for sample image data was developed using the resolved absorbance values for RGB channels (standardized via self-scaling). The multiple linear regression for the CCS analysis showed a value of R2 = 097 and for the enzymatic activity R2 = 0.96, signaling that the RGB channels increased in intensity according to the increase of the concentration of CCS and lactoperoxidase in the samples, then lactoperoxidase imaging is indicative of subclinical mastitis in cattle.