Abstract #216

Section: Dairy Foods
Session: Dairy Foods I: Dairy Products
Format: Oral
Day/Time: Monday 3:30 PM–3:45 PM
Location: 331
# 216
Evaluation of electrical bioimpedance spectroscopy in estimate the milk composition, SCC, and milk ethanol stability—Preliminary results.
C. M. M. R. Martins*1, E. A. Veiga2, D. C. M. Fonseca1, B. G. Alves1, M. V. Santos1, 1Department of Animal Nutrition and Production, School of Veterinary Medicine and Animal Science, University of São Paulo, Pirassununga, São Paulo, Brazil, 2Bionexus Tecnology, Chapecó, Brazil.

Electrical bioimpedance spectroscopy (EBS) is a fast, easy and low-cost methodology used for measurement of electrical properties of biological materials and quality of food. The present study aimed to evaluate an automatic equipment based on EBS for measurement of milk quality. Individual composite milk samples from 2 daily milkings were collected from 25 Holstein cows averaging (mean ± SD) 25 ± 5 kg of milk/d and 150 ± 6 DIM. Milk samples were stored for 24 h at 5°C. Milk ethanol stability (MES) was performed at following concentrations (vol/vol): 64, 68, 70, 72, 74, 76, 78, 80, 82 and 84% ethanol. Results were expressed by the lowest concentration of ethanol in which coagulation occurred. The EBS analysis made by Milkspec FS317 (Bionexus, Brazil) required only 20 mL of raw milk and the bioimpedance spectra were obtained in less than a minute. The Bionexus InterCurve EBS Model software for curve fitting calculated the parameters list fitted to the impedance model and temperature compensation. At the end of this first step, milk samples were heated at 37°C in attempt to study the efficiency of milk heat on accuracy of EBS predicted milk quality. Fat, lactose and protein were determined by infrared absorption. The mean spectra parameter related with EBS curve behavior fitted by Bionexus InterCurve were obtained and associated with MES, milk composition and SCC by correlation and regression model. MES was better predicted using cooled samples by the mean of bioimpedance expressed in ohms (R2 = 0.49; P < 0.0001; RMSE = 4.31). Milk composition was better predicted at 37°C, as it was possible estimated the milk concentrations of lactose (R2 = 0.63; P < 0.0001; RMSE = 0.2790), fat (R2 = 0.23; P = 0.0292; RMSE = 0.837) and log 10 of SCC (R2 = 0.6; P < 0.0001; RMSE = 0.339) using parameters of EBS estimated from Bionexus InterCurve EBS Model. Milk protein were not associated with EBS. This preliminary results suggest that an automatic equipment based on EBS may be used in dairy farms and dairy plants to estimate milk quality parameters, with easier operating procedures and lower cost than traditional methods.

Key Words: electrical bioimpedance spectroscopy, mastitis detection, milk quality