Hyperspectral Imaging Is a Cutting Edge Food Safety Technology

The advent of widely accessible machine learning methods has also brought a new and powerful set of tools to HSI pathogen detection. To make use of the abundance of data rendered by HSI, a number of image processing algorithms have been developed over the years, with more created all the time. These mathematical techniques, combined with intelligent HSI microscopy, aid users in interpreting the data with speed and accuracy.

In comparison to HSI, other traditional quality assurance systems pose additional specific limitations in regard to food safety and pathogen detection. X-rays, which are prohibitively expensive and only focus on detecting foreign objects, can be difficult to maintain and calibrate. While metal detectors are more affordable, they generally only catch metals with strong magnetic fields like iron. Unfortunately, many materials, including copper, aluminum, plastics, wood, and feces, can slip through undetected.

Conventional quality assurance systems also rely on human subjectivity, which may shift subtly from day to day or even hour to hour. While those in charge of monitoring in-line quality and food safety are trying their best, the naked eye and human brain can be erratic. Tired or distracted people may judge quality in different ways, leading to inconsistent standards that can negatively affect both the food processor and consumers.

Using HSI for Pathogen Detection

Compared with current conventional techniques, HSI can immediately provide tangible benefits for the food industry, especially when it comes to quality assurance in the food supply chain. HSI solutions provide benefits across three specific elements.

First, the intelligent hyperspectral microscopy employed within HSI combines the analytical benefits of conventional spectroscopy with digital microscopy imaging to provide high resolution spectral and spatial information that enables the spectral identification of every pixel throughout bacterial cell images. Dark field illumination microscopy is also utilized to negate the need for staining or special reagent growth media. This vastly shortens the time needed to identify potentially harmful pathogens in line and enables them to be identified on premises.

Second, HSI utilizes machine learning to constantly improve its image processing capabilities, helping food processors better monitor and control the quality of their food products. As described above, the hyperspectral imager hardware returns a raw data cube, which represents the spectral information for each pixel in the image. By using special software analysis and calibration, HSI products continue to improve discernibility as larger sample sizes and additional data are processed. HSI is built upon a set of machine learning tools specially trained to identify the dominant spectral and spatial signatures in harmful food pathogens and label them for intuitive identification. This helps provide instant feedback for food processors, who can then implement new systems with this instant data. For example, if a harmful pathogen is detected, lines will be stopped automatically so the issue can be resolved immediately.

Finally, HSI utilizes automated detection to identify pathogen cells. Traditional processes for correctly identifying pathogen cells can be involved and time-consuming. HSI can automate the capture of spectral data cubes from a sparsely populated field of view. In the case of low cell concentrations, the process automatically searches for cells using a targeting algorithm. The system can then enable the rapid identification of pathogens on site in less than four hours, thereby reducing testing time exponentially.

Additional In-Line Applications

In addition to food pathogen detection, multiple food characteristics can be measured simultaneously with HSI, including color, moisture levels, fat content, and protein levels. This level of nuance gives details of chemical and structural composition not discernible to the naked eye, while providing added information that can be used by manufacturers for improved safety and quality assurance in a number of areas.