A Voting-Based Ensemble Deep Learning Method Focused on Multi-Step Prediction of Food Safety Risk Levels: Applications in Hazard Analysis of Heavy Metals in Grain Processing Products.

Zuzheng Wang,Zheng Wang,Qingchuan Zhang,Minke Zou,Zhixiang Wu,Xin Wen,Yuanzhang Li

doi:10.3390/foods11060823

Abstract

Grain processing products constitute an essential component of the human diet and are among the main sources of heavy metal intake. Therefore, a systematic assessment of risk factors and early-warning systems are vital to control heavy metal hazards in grain processing products. In this study, we established a risk assessment model to systematically analyze heavy metal hazards and combined the model with the K-means++ algorithm to perform risk level classification. We then employed deep learning models to conduct a multi-step prediction of risk levels, providing an early warning of food safety risks. By introducing a voting-ensemble technique, the accuracy of the prediction model was improved. The results indicated that the proposed model was superior to other models, exhibiting the overall accuracy of 90.47% in the 7-day prediction and thus satisfying the basic requirement of the food supervision department. This study provides a novel early-warning model for the systematic assessment of the risk level and further allows the development of targeted regulatory strategies to improve supervision efficiency.

Highlights

Appropriate food safety assessment methods have to be implemented to determine the effect of heavy metal contamination on the safety of grain processing products
Considering the need to integrate the heavy metal hazards of grain processing products with environmental and health factors, we introduced food safety risk assessment indexes—target cancer risk (TCR), target hazard quotient (THQ), and Nemerow integrated pollution index (NIPI)—to comprehensively measure the heavy metal hazard in grain processing products and used them as inputs in early-warning models
Recent studies have focused on conducting a risk assessment of food contaminants, in addition to a food safety risk evaluation based on the calculation results for the detection samples via analytic hierarchy process based on entropy weight (AHP-EW), in fields implementing early-warning systems

Summary

Introduction

Food quality and safety issues have drawn wide interest with the continuous development of the social economy [1]. In China, more stringent regulatory measures have been implemented by the government. Despite these efforts, food safety incidents still arise [3]. Heavy metal deposition in agroecosystems has recently increased, in northern China, posing serious risks to crop safety and human health via the food chain [6]. Appropriate food safety assessment methods have to be implemented to determine the effect of heavy metal contamination on the safety of grain processing products. Food quality and safety are closely related to the health and living standard of individuals, and the risk assessment of food quality and safety bears considerable social significance [11]. Several studies on risk assessment [11–13] have applied AHP-EW to determine objective food safety risk values as inputs in early-warning models. Niu et al [14] established dietary exposure models, which are typically used to assess the carcinogenic and non-carcinogenic risks of children and adults after metal exposure [15], allowing for a comprehensive assessment of the health risks in vegetables and providing scientific and comprehensive support for risk assessments

Methods

Results

Discussion

Conclusion