SBIR-STTR Award

With increased seafood imports and limited monitoring, fraud and deception in seafood marketing is a growing safety and quality concern. The flesh of many fish species is similar in taste and texture and, therefore, it is difficult to identify species in fillet form. SafetySpect is developing a tabletop system incorporating multimode spectroscopy for real-time quality, adulteration and traceability (QAT) assessment of fish portions. The fish portion will be placed on the scanner plate for immediate analysis. The multimode spectroscopy measurement will be analyzed using algorithms developed using artificial intelligence methods and will display the fish species, quality and nutritional content in seconds. Phase 1 of this project will determine the feasibility of extending existing multimode spectroscopic data analysis for assessment of fish species and quality from 10 to 50 species commonly consumed in the US. We will collect data on fish species, quality, and nutrient content and combine these with spectral data using four reference spectroscopic systems. We will demonstrate scalability and evaluate accuracy of machine learning classification algorithms for the assessment of fish quality, nutrient content, and species authentication. We will determine the design/market requirements and product characteristics for implementing the QAT system in the seafood supply chain."

A goal of the “Illegal, Unreported, and Unregulated Fishing Enforcement Act” is to prevent illegally harvested fish from entering our ports and market and to combat illegal, unreported, and unregulated (IUU) fishing. Seafood is highly vulnerable to fraud due to factors such as the similar appearance of many species, variation in prices, complex supply chains, and challenges with supply and demand. Instances of seafood fraud are frequently reported; however, many incidents go undetected and the full extent of seafood fraud is difficult to determine. Existing rapid DNA testing can only detect a single species with each assay. There is a need for technology to rapidly detect mislabeling. The SafetySpect Quality, Adulteration and Traceability (QAT) system delivers innovation in the management of seafood supply chains. QAT provides rapid species identification, ability to incorporate many species (unlike rapid DNA analysis), bypasses sample preparation, and reduces the time and cost of species identification. QAT uses multimode spectroscopy, combining reflectance and fluorescence-based spectral analysis and a fusion AI classification algorithm. During Phase II, we will design and build handheld multimode spectroscopy systems, test them at several nodes in the seafood supply chain and expand the spectral dataset up to 200 fish and shellfish species.