首页|University of Sydney Reports Findings in Machine Learning (Comparison of a machi ne learning model with a conventional rule-based selective dry cow therapy algor ithm for detection of intramammary infections)
University of Sydney Reports Findings in Machine Learning (Comparison of a machi ne learning model with a conventional rule-based selective dry cow therapy algor ithm for detection of intramammary infections)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News – New research on Machine Learning is the subject o f a report. According to news originating from Camden, Australia, by NewsRx corr espondents, research stated, “We trained machine learning models to identify int ramammary infections (IMI) in late lactation cows at dry-off to guide antibiotic treatment, and compared their performance to a rule-based algorithm that is cur rently used on dairy farms in the US. We conducted an observational test-charact eristics study using a data set of 3,645 cows approaching dry-off from 68 US dai ry herds.” Our news journalists obtained a quote from the research from the University of S ydney, “The outcome variables of interest were cow-level IMI caused by all patho gens, major pathogens, and Streptococcus and Strep-like organisms (SSLO), which were determined using aerobic culture of aseptic quarter-milk samples and identi fication of isolates using MALDI-TOF. Individual cow records were extracted from the farm software to create 53 feature variables at the cow and 39 at the herd- level which were derived from cowlevel descriptive data, records of clinical ma stitis events, results from routine testing of milk for volume and concentration s of somatic cell count (SCC), fat, and protein. ML algorithms evaluated were lo gistic regression, decision tree, random forest, light gradient-boosting machine , naive bayes, and neural networks. For comparison, cows were also classified ac cording to a conventional rule-based algorithm that considered a cow as high ris k for IMI if she had at one or more high SCC (>200,000 c ells/ml) tests or 2 cases of clinical mastitis during the lactation of enrollmen t. Area under the curve (AUC) and Youden’s index were used to compare models, in addition to binary classification metrics, including sensitivity, specificity, and predictive values. ML models had slightly higher AUC and Youden’s index valu es than the rule-based algorithm for all IMI outcomes of interest. However, thes e improvements in prediction accuracy were substantially less than what we had c onsidered necessary for the technology to be a worthwhile alternative to the rul e-based algorithm. Therefore, evidence is lacking to support the wholesale use o f ML-guided selective dry cow therapy at the moment.”
CamdenAustraliaAustralia and New Zea landAlgorithmsCyborgsEmerging TechnologiesMachine LearningTherapy
NETL
NSTL
