Milk is one of the most valuable products in terms of nutritional value and balanced composition. About a third of the entire herd of cows gets sick with mastitis, while milk from such cows must be disposed of due to the presence of pathogenic microorganisms and their toxins. Intensification and robotization of various areas of the agro-industrial complex requires modern systems for assessing the quality of raw materials and food products. One of the possible options for detecting mastitis milk is the use of sensor technologies. A method is proposed for detecting milk from cows with mastitis by analyzing the gas phase above milk samples using chemical sensors, including the subclinical course of the disease with a pathogenic microorganism content in milk of no more than 1000 CFU / ml, and differentiating such samples from milk with a high level of coliform bacteria and a general level of contamination, as well as with non-vital forms of pathogenic microorganisms. The work used piezoelectric quartz sensors with polycomposite coatings to identify informative sensor output data related to the presence of pathogenic microorganisms and their toxins, milk samples were analyzed in parallel using microbiological and molecular genetic methods. The work demonstrated approaches to determining Staphylococcus, Streptococcus, Klebsiella spp. in milk from mastitis cows at a level of 1000-10000 CFU/ml with high sensitivity due to variation in the preparation of milk samples before measurement. Thus, the use of gas sensors will allow obtaining information on the microbiological safety of milk within three hours.