Table 3.

Effects of artificial intelligence for the application of food allergy.

Year	Author	Specific Artificial Intelligence	Reference	Diagnose/prediction/treatment/management	efficiency
2013	Dimitrov, et al	KNN	57	Prediction	SE: 94%; SP: 94%; F1-Score: 94%
2019	Metwally, et al	Long Short-Term Memory (LSTM) networks	56	Prediction	AUROC: 0.69; MCC: 0.40
2023	Xin-Xin Yu, et al	Quantitative structure-activity Relationship (QSAR) models	105	Prediction	RMSE: 0.2375
2024	Landau, et al	Random Forest Regression	58	Prediction	AUROC: 0.8
2019	Alag, et al	A Java-based machine learning toolkit	106	Diagnose	cg06410630 and cg06669701 are higher for food-allergic patients.
2021	Kuniyoshi, et al	LR; SVM; XGBoost	60	Diagnose; prediction	LR: AUROC (82%); SE (70%); SP (73%); AC (72%) SVM: AUROC (83%); SE (68%); SP (74%); AC (72%) XGB: AUROC (63%); SE (51%); SP (66%); AC (59%)
2023	Lucy, et al	NLP	62	Diagnose	AUROC: LR (0.84); Passive-aggressive (83.0%)
2022	Pradana-López, et al	CNN	65	Treatment; management	AC: 99.1%
2024	Liu, et al	Artificial Intelligence	66	Treatment	Screen out three potential therapeutic targets: IL-5, PTAFR and RNF19B.