Smartphone app with cloud-based machine learning and visual sensor array for instrument-free detection of species, geographic origin, and quality marker in functional food Gancao

Abstract

Glycyrrhizae Radix et Rhizoma (Gancao) is a functional food whose quality varies significantly owing to the genetic and geographical factors, and is often used by unscrupulous merchants as a substitute for profit. This study developed a rapid identification method for Gancao that determined the species, geographical origin, and main quality marker using a six-channel visual array sensor. This sensor detected the color changes resulting from the competitive coordination with metal ions and color-changing organic dyes. The combination of the array sensor with a random forest (RF) algorithm achieved 100% accuracy in identifying Gancao. Additionally, the smartphone app enabled the rapid and intelligent identification of Gancao within 1 min. The identification accuracy was 91.43%, and the prediction error of two quality markers was less than 13%. This research provides a foundation for rapid and reliable assessment of Gancao quality, enabling the industry to combat fraudulent practices effectively.