Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Environmental Microbial & Food Safety Laboratory » Research » Research Project #440143

Research Project: Machine Learning-Enabled Novel Pathogen Detection Platform for Nondestructive Supply chain Surveillance

Location: Environmental Microbial & Food Safety Laboratory

Project Number: 8042-32420-009-013-S
Project Type: Non-Assistance Cooperative Agreement

Start Date: May 1, 2021
End Date: Apr 30, 2024

Objective:
This project seeks to develop machine learning-enabled pathogen detection platforms using nano technology. Specifically, we will develop nondestructive sensing platforms for foodborne human pathogens using bioinspired nanomaterials including photonic crystals and other nontoxic or food grade chromogenic dyes; and 2) develop machine learning algorithms to enable pathogen detection in the presence of natural background microbiome on food matrices.

Approach:
The proposed work will extend our paper chromogenic array (PCA) platform for multiplex viable pathogen detection using novel nontoxic or food grade dyes as sensing elements. Specifically, this new approach will investigate the potential of bio-inspired photonic crystals, derived from natural pigments of color-changing animals, like chameleons and cephalopods. Photonic crystals produce coloration largely based on diffraction which occurs when light reaches an object or slit on the same size order of the wavelength of light and bends around it. The most familiar natural material with structural color is the opal, where the dynamic iridescent colors come from periodically ordered arrays of monodispersed silica (SiO2) spheres with diameters on the sub micrometer scale. Biological photonic crystals are abundant in nature and are often observed as highly ordered nanostructure arrays that can generate structural coloration, which plays an important role in animals with active camouflage via sub-micron melanosomes. As chromogenic sensing elements, photonic crystals are responsive to various stimuli, including mechanical deformation, temperature, humidity, pH, electric field, magnetic field, ions, molecules, DNAs, and volatile organic compounds (VOCs). Photonic crystal sensors prepared using patterned arrays of nanoparticles have great potential as continuous safety monitoring of food products. The sensors proposed here will incorporate ordered nanostructures, typically in the size order of the wavelengths of light they interact with, to generate constructive and destructive interference which allows for the reflection of different wavelengths in the visible spectrum. This reflection can be facilitated by varying material’s refractive index. By controlling the size of the particles, their packing density, and the thickness of the patterned films, we can leverage the natural light reflecting and scattering properties of these materials as smart optical sensors. We hypothesize that organic vapors produced by microorganisms can change the average refractive index of the photonic crystals and induce a global colorimetric change by adsorption to the nanoparticles. In this configuration, the sensors can undergo a specific colorimetric response upon detection of target volatile organic compounds (VOC) that are indicative of microbial pathogens in the headspace. The photonic crystal sensor array will be coupled with advanced machine learning (ML) algorithms for multiple objectives, including differentiate VOC categories using photonic crystals with similar physiochemical properties; differentiate pathogen targets and achieve multiplex detection using photonic crystal arrays. The system will be validated using fresh produce models (such as lettuce, spinach, cantaloupe, etc.) against prominent produce-borne pathogens, like E. coli O157:H7, Salmonella spp., Listeria spp., etc. in the presence of typical background microbiome.