The invention discloses a method and
system for prenatal early warning of cow
ketosis by using an electronic
nose, and the method comprises the steps: S1, collecting cow excrement samples 21-14 days before delivery, synchronously collecting blood samples 3-14 days after delivery, detecting the concentration of beta-
hydroxybutyric acid, and dividing a healthy group and a
ketosis group according to a preset beta-
hydroxybutyric acid concentration threshold; s2, based on the excrement samples of the healthy group and the
ketosis group,
odor sensor response values of the excrement samples are collected through an electronic
nose, an
odor mark map is obtained, and a plurality of sensor arrays with significant differences are obtained through screening; s3, based on the data of the plurality of sensor arrays with significant differences, an RF (
Random Forest)
algorithm is adopted to construct an
early warning model; and S4, inputting to-be-detected data into the
early warning model, and outputting a ketosis risk prediction result. According to the method, antenatal early warning is achieved through non-invasive excrement VOCs analysis, the method has the advantages of being convenient to operate, efficient in detection, economical,
environmentally friendly and the like, and a reliable early health monitoring scheme is provided for a large-scale
pasture.