USDA

United States Department of Agriculture 

USDA

Smart Vision and AI-driven Peach Thinning (A-EYE)

Duration: 2022-2023

Role: PI


Research Objective: This projects is a SBIR Phase I grant subcontracted to USC by Advent Innovation to perform one specific task which is Digitization of human knowledgebase for guiding peach thinning. More details to follow as project progresses. 

TOXIMAP to evaluate the incidence of mycotoxins in US crop fields

Duration: 2017-2022

Role: Co-PI


Research Objectives: Aflatoxin is a carcinogenic toxin produced by Aspergillus-family fungi occurring in soil and decaying vegetation, which can contaminate corn along with other relevant US crops like peanuts, treenuts and cotton before harvest and/or during storage. Aflatoxin causes liver cancer in humans and in a variety of animal species, and it has been associated with childhood stunting; an important public health issue that increases vulnerability to infectious diseases and cognitive impairments. It is estimated that 5 billion people worldwide are at risk of aflatoxin exposure. Prediction and control of aflatoxin contamination is a fundamental challenge for US grain industry, poultry producers, and makers of dairy products. The production of aflatoxin is highly dependent on environmental conditions and it is projected that environmental perturbations due to climate change will result in a significant increase in aflatoxin contamination incidents, further aggravating its economic impact. The lack of a systematic approach to determine the distribution of aflatoxin occurrence before harvest adversely affects the grain industry. The goal of this project was to develop a general predictive modeling framework for calculating aflatoxin occurrence in US crop fields before harvest, and package this knowledge in a user-friendly predictive web/mobile interface for generating nation-wide and real-time aflatoxin hazard maps. This has the potential to change certain behaviors in crop management to improve food safety (lower levels of aflatoxin) such as providing valuable information of when and where to test for aflatoxin before harvest, reducing the amount of chemicals used to grow the crops, determining the best time to harvest, as well as indicating the cornfield areas with the highest contamination levels for grain segregation. The mathematical foundation and the proposed software tool are generally applicable to any type of mycotoxin and crop. While hazard maps for aflatoxin and other mycotoxin occurrence are important to crop producers, they are also of interest to consumers, regulators, and insurers to enhance their decision-making process. Furthermore, the presence of mycotoxins in agricultural run-offs is a critical issue in environmental sustainability and ecological physiology of various land and aquatic ecosystems. Hence, the approach to be developed will be of relevance to a broader community beyond the agricultural community. Example of a tentative Toximap.