Scholarlink

<p>The study involved preparing edible coatings using <i>Alyssum homolocarpum</i> seed mucilage (AHSM) and the cell-free supernatant (CFS) of <i>Lactiplantibacillus pentosus</i> SM1 to examine their impact on the quality and shelf-life of fresh beef during refrigerated storage. Uncoated and coated beef samples were regularly assessed during storage at 4&#xa0;°C for microbial colony growth (total viable count-TVC, the psychrotrophic count-PTC, and fungi), lipid oxidation (thiobarbituric acid reactive substances-TBA and peroxide value-PV), pH, moisture content, texture, and sensory characteristics. Also, in this study, Gaussian Process Regression (GPR) and Support Vector Machine (SVM) models were used to predict some of laboratory parameters. The coating treatments notably slowed down the beef oxidation process, with TBA values of 0.73 and 0.35&#xa0;mg MDA/kg and PV values of 5.92 and 2.71 meq O2/kg for uncoated and AHSM + 2%CFS coated samples, respectively. Similarly, for AHSM + 2%CFS coated samples, the microbial counts for TVC, PTC, and fungi were 5.05, 3.17, and 2.45 log CFU/g, which were lower compared to the counts of uncoated, AHSM, and AHSM + 1%CFS coated samples. Furthermore, the texture and sensory evaluations indicated enhanced characteristics in the beef coated with AHSM + 2%CFS during the storage period. The discoveries could form the basis for creating AHSM + CFS as a bacterial-resistant coating for storing chilled beef. The results of modelling indicated that GPR has the highest accuracy than SVM model. The lowest and highest MAPE by using GPR model was related to moisture (0.43%) and TBA (16.13%). This makes GPR highly effective for small datasets where traditional models might struggle. Additionally, GPR’s ability to interpolate and smooth data helps in generating reliable predictions even with sparse data points.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Postbiotic Lactiplantibacillus pentosus SM1-infused bioactive packaging: production, characterization, application, and AI modeling of beef slice preservation

  • 期刊论文

  • 发表日期：

    2025年11月3日

查看全文
  • Maryam Jooyandeh Kar,
  • Behrooz Alizadeh Behbahani,
  • Alireza Vasiee,
  • Hossein Jooyandeh,
  • Morteza Taki

摘要

The study involved preparing edible coatings using Alyssum homolocarpum seed mucilage (AHSM) and the cell-free supernatant (CFS) of Lactiplantibacillus pentosus SM1 to examine their impact on the quality and shelf-life of fresh beef during refrigerated storage. Uncoated and coated beef samples were regularly assessed during storage at 4 °C for microbial colony growth (total viable count-TVC, the psychrotrophic count-PTC, and fungi), lipid oxidation (thiobarbituric acid reactive substances-TBA and peroxide value-PV), pH, moisture content, texture, and sensory characteristics. Also, in this study, Gaussian Process Regression (GPR) and Support Vector Machine (SVM) models were used to predict some of laboratory parameters. The coating treatments notably slowed down the beef oxidation process, with TBA values of 0.73 and 0.35 mg MDA/kg and PV values of 5.92 and 2.71 meq O2/kg for uncoated and AHSM + 2%CFS coated samples, respectively. Similarly, for AHSM + 2%CFS coated samples, the microbial counts for TVC, PTC, and fungi were 5.05, 3.17, and 2.45 log CFU/g, which were lower compared to the counts of uncoated, AHSM, and AHSM + 1%CFS coated samples. Furthermore, the texture and sensory evaluations indicated enhanced characteristics in the beef coated with AHSM + 2%CFS during the storage period. The discoveries could form the basis for creating AHSM + CFS as a bacterial-resistant coating for storing chilled beef. The results of modelling indicated that GPR has the highest accuracy than SVM model. The lowest and highest MAPE by using GPR model was related to moisture (0.43%) and TBA (16.13%). This makes GPR highly effective for small datasets where traditional models might struggle. Additionally, GPR’s ability to interpolate and smooth data helps in generating reliable predictions even with sparse data points.