Detection of bacteria in food packaging with laser driven spectral analyses and machine learning

Abstract

Abstract The rapid and non- destructive detection and identification of bacteria is an important current topic in research. Therefore laser spectroscopy in combination with advanced analytical methods is a promising technology. This work is dedicated to the proof of the concept of detecting bacteria in food samples through the packaging with laser driven spectroscopy. Therefore a literature research identified the Raman spectroscopy as a suitable technology for detection and identification of bacteria with laser spectroscopy. To check if the Raman spectroscopy detect bacteria through the packaging, real world vacuum bags have been examined with own measurements. The transmittance, the fluorescence characteristics have been measured and basic machine learning models have been tested on measurements data of a bacterial suspension in a vacuum bag. The results of the measurements and the metrics of the ML models have shown, that the detection through the packaging should be possible with Raman spectroscopy.