Metodologia para identificação de crescimento microbiano em tempo real empregando técnica de medição não invasiva

Abstract

This thesis discusses the creation of a new methodology for the real-time, non-invasive monitoring of microbial growth in food microbiology. In the face of the complexity and time required by conventional microbiological analysis, the study proposes a strategy to assess the dielectric behavior of a sample in a standard 9 mL test tube, detecting alterations resulting from biochemical processes through variations in the resonance frequency of a resonant system. The research also included the development of a device that controls the sample’s temperature and promotes agitation, providing ideal conditions for microbial growth. The results indicate a significant reduction in microbiological analysis time, with the proposed methodology simplifying sample preparation and reducing the number of necessary steps. The Smooth-Spline technique proved particularly suitable for bacterial growth analysis, with the resonance frequency-based method showing to be faster than conventional turbidimetry and nephelometry methods. The results also suggest that the proposed sensor could effectively classify milk quality, allowing for the distinction between Class A and Class B milk, in accordance with Normative Instruction No. 76 of November 26, 2018.