Combining AI Tools with Non-Destructive Technologies for Crop-Based Food Safety: A Comprehensive Review

Abstract

On a global scale, food safety and security aspects entail consideration throughout the farm-to-fork continuum, considering food's supply chain. Generally, the agrifood system is a multiplex network of interconnected features and processes, with a hard predictive rate, where maintaining the food's safety is an indispensable element and is part of the Sustainable Development Goals (SDGs). It has led the scientific community to develop advanced applied analytical methods, such as machine learning (ML) and deep learning (DL) techniques applied for assessing foodborne diseases. The main objective of this paper is to contribute to the development of the consensus version of ongoing research about the application of Artificial Intelligence (AI) tools in the domain of food-crop safety from an analytical point of view. Writing a comprehensive review for a more specific topic can also be challenging, especially when searching within the literature. To our knowledge, this review is the first to address this issue. This work consisted of conducting a unique and exhaustive study of the literature, using our TriScope Keywords-based Synthesis methodology. All available literature related to our topic was investigated according to our criteria of inclusion and exclusion. The final count of data papers was subject to deep reading and analysis to extract the necessary information to answer our research questions. Although many studies have been conducted, limited attention has been paid to outlining the applications of AI tools combined with analytical strategies for crop-based food safety specifically.

Keywords: chemometrics; food contaminants; food processes; machine learning; spectroscopy; sustainability.

Figure 1

Generalized representation of our MultiScope Keywords-based Synthesis methodology.

Figure 2

The general workflow of our developed TriScope Keywords-based Synthesis for conducting this comprehensive review.

Figure 3

Data collection process, using API for the Scopus databse and manual searching for Web of Science and IEEE Xplore databases.

Figure 4

Diagram illustrating the inclusion–exclusion process and criteria of selection.

Figure 5

A map of all our collected papers. Each node represents a paper with its first author and year of publication. The size of a node represents the number of citations. Please refer to Table 3 for more details.

Figure 6

Count of published papers per publisher name.

Figure 7

A table chart depicting the number of papers published per journal.

Figure 8

A diagram representing the chronological (from 2001 to 2022) distribution of the resulting articles. Each data-point represents an article, the colors are according to which field of study, and the lines are related to the decision-making objective Mehl, 2001 [47]; Irudayaraj, 2002 [48]; Yang, 2003 [49]; Gupta, 2005 [50]; Gupta, 2006 [51]; He, 2008 [55]; A. Scarlatos, 2008 [54]; Siripatrawan, 2008 [56]; Stöckel, 2010 [57]; Günes, 2013 [60]; Shapaval, 2013 [59]; Geng, 2017 [66]; Y.Shen, 2017 [67]; Lasch, 2018 [68]; Guo, 2019 [72]; Kaushik, 2019 [69]; Liu, 2019 [74]; Öner, 2019 [71]; Sun, 2019 [75]; Wan-dan, 2019 [73]; Bertania, 2020 [85]; Le, 2020 [77]; Sahu, 2020 [80]; Shen, 2020 [84]; Wange, 2020 [78]; Weng, 2020 [79]; Wu, 2020 [83]; Gonzalez, 2021 [93]; Guo, 2021 [94]; Li, 2021 [95]; Magnus, 2021 [91]; Nie, 2021 [101]; Rahi, 2021 [86]; Ricci, 2021 [97]; Vakilian, 2021 [96]; Wang, 2021 [92]; Xu, 2021 [90]; Yan, 2021 [99]; Yin, 2021 [98]; Adejimi, 2022 [109]; Bowler, 2022 [107]; Cordovana, 2022 [110]; Kim, 2022 [108]; Manthou, 2022 [115]; Rady, 2022 [106]; Wang, 2022 [111]; Yan, 2022 [104]; Yang, 2022 [102].

Figure 9

A map highlighting subject-related words and concepts that were strongly present within the literature, such as Spectroscopy (Graphical design generated using free-online tool Wordart https://wordart.com/create, accessed on 23 August 2023).

Figure 10

Information and data workflow according to case studies within our collected papers.

Figure 11

Statistics about evaluation metrics and the use of K-fold cross-validation in the research papers. (a) Number of times each evaluation metric is used in the 49 research papers. (b) Count of evaluation metrics used per research paper. (c) Percentage of research papers that uses K-fold cross-validation for model validation. (d) The number of folds used in K-fold.