A Machine–Learning Approach Reveals That Bacterial Spore Levels in Organic Bulk Tank Milk are Dependent on Farm Characteristics and Meteorological Factors
IF 2.1 4区 农林科学Q3 BIOTECHNOLOGY & APPLIED MICROBIOLOGY
Chenhao Qian, Renee T. Lee, Rachel L. Weachock, Martin Wiedmann, Nicole H. Martin
{"title":"A Machine–Learning Approach Reveals That Bacterial Spore Levels in Organic Bulk Tank Milk are Dependent on Farm Characteristics and Meteorological Factors","authors":"Chenhao Qian, Renee T. Lee, Rachel L. Weachock, Martin Wiedmann, Nicole H. Martin","doi":"10.1016/j.jfp.2025.100477","DOIUrl":null,"url":null,"abstract":"<div><div>Bacterial spores in raw milk can lead to quality issues in milk and milk-derived products. As these spores originate from farm environments, it is important to understand the contributions of farm-level factors to spore levels. This study aimed to investigate the impact of farm management practices and meteorological factors on levels of different spore types in organic raw milk using machine learning models. Raw milk from certified organic dairy farms (<em>n</em> = 102) located across 11 states was collected 6 times over a year and tested for standard plate count, psychrotolerant spore count, mesophilic spore count, thermophilic spore count, and butyric acid bacteria. At each sampling date, a survey about farm management practices was collected and meteorological factors were obtained on the date of sampling as well as 1, 2, and 3 days prior. The dataset was stratified separately based on the use of a parlor for milking, number of years since organic certification, and pasture time into subdatasets to address confounders. We constructed random forest regression models to predict log<sub>10</sub> mesophilic spore count, log<sub>10</sub> thermophilic spore count, and log<sub>10</sub> butyric acid bacteria’s most probable number as well as a random forest classification model to classify the presence of psychrotolerant spores in each raw milk sample. The summary statistics showed that spore levels vary considerably between certified organic farms but were only slightly higher than those from conventional farms in previous longitudinal studies. The variable importance plots from the models suggest that herd size, certification year, employee-related variables, clipping and flaming udders are important for the spore levels in organic raw milk. The small effects of these variables as shown in partial dependence plots suggest a need for individualized risk-based approach to manage spore levels. Incorporating novel data streams has the potential to enhance the performance of the model as a real-time monitoring tool.</div></div>","PeriodicalId":15903,"journal":{"name":"Journal of food protection","volume":"88 5","pages":"Article 100477"},"PeriodicalIF":2.1000,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of food protection","FirstCategoryId":"97","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0362028X25000298","RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"BIOTECHNOLOGY & APPLIED MICROBIOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Bacterial spores in raw milk can lead to quality issues in milk and milk-derived products. As these spores originate from farm environments, it is important to understand the contributions of farm-level factors to spore levels. This study aimed to investigate the impact of farm management practices and meteorological factors on levels of different spore types in organic raw milk using machine learning models. Raw milk from certified organic dairy farms (n = 102) located across 11 states was collected 6 times over a year and tested for standard plate count, psychrotolerant spore count, mesophilic spore count, thermophilic spore count, and butyric acid bacteria. At each sampling date, a survey about farm management practices was collected and meteorological factors were obtained on the date of sampling as well as 1, 2, and 3 days prior. The dataset was stratified separately based on the use of a parlor for milking, number of years since organic certification, and pasture time into subdatasets to address confounders. We constructed random forest regression models to predict log10 mesophilic spore count, log10 thermophilic spore count, and log10 butyric acid bacteria’s most probable number as well as a random forest classification model to classify the presence of psychrotolerant spores in each raw milk sample. The summary statistics showed that spore levels vary considerably between certified organic farms but were only slightly higher than those from conventional farms in previous longitudinal studies. The variable importance plots from the models suggest that herd size, certification year, employee-related variables, clipping and flaming udders are important for the spore levels in organic raw milk. The small effects of these variables as shown in partial dependence plots suggest a need for individualized risk-based approach to manage spore levels. Incorporating novel data streams has the potential to enhance the performance of the model as a real-time monitoring tool.
期刊介绍:
The Journal of Food Protection® (JFP) is an international, monthly scientific journal in the English language published by the International Association for Food Protection (IAFP). JFP publishes research and review articles on all aspects of food protection and safety. Major emphases of JFP are placed on studies dealing with:
Tracking, detecting (including traditional, molecular, and real-time), inactivating, and controlling food-related hazards, including microorganisms (including antibiotic resistance), microbial (mycotoxins, seafood toxins) and non-microbial toxins (heavy metals, pesticides, veterinary drug residues, migrants from food packaging, and processing contaminants), allergens and pests (insects, rodents) in human food, pet food and animal feed throughout the food chain;
Microbiological food quality and traditional/novel methods to assay microbiological food quality;
Prevention of food-related hazards and food spoilage through food preservatives and thermal/non-thermal processes, including process validation;
Food fermentations and food-related probiotics;
Safe food handling practices during pre-harvest, harvest, post-harvest, distribution and consumption, including food safety education for retailers, foodservice, and consumers;
Risk assessments for food-related hazards;
Economic impact of food-related hazards, foodborne illness, food loss, food spoilage, and adulterated foods;
Food fraud, food authentication, food defense, and foodborne disease outbreak investigations.