Table 2.
Summary of the studies of artificial intelligence (AI) in early intervention of rheumatoid arthritis (RA), including authors, data types, methods, and main findings of the same category
| Main findings | Data types | Methods | Author (Month/Year) |
|---|---|---|---|
| ML techniques have demonstrated their value in the field of early intervention in RA by analyzing EHRs, clinical trial data and claims data. These | EHRs, claims data, clinical cohort, RCTs | DL | Norgeot et al.[38] (Mar. 2019) |
| studies have used ML models to predict clinical | NLP | Spencer et al.[42] (Nov. 2021) | |
| outcomes, estimate disease activity scores, predict treatment response, identify predictors of severe COVID-19 outcomes, and cluster comorbidities in RA patients. These findings | |||
| highlight the potential of ML to improve early intervention strategies for RA by leveraging | XGBoost, SVM | Morid et al.[43] (Jul. 2021) | |
| multiple healthcare data sources. | Random forests | Johansson et al.[44] (May. 2021) | |
| Logistic regression | Burns et al.[45] (Nov. 2022) | ||
| Hierarchical clustering, factor analysis, k-means clustering, and network analysis | Crowson et al.[46] (Feb. 2023) | ||
| Factor analysis | England et al.[47] (Feb. 2023) | ||
| Linear regression, lasso and ridge, SVM, random forest, and XGBoost | Koo et al.[48] (Jul. 2021) | ||
| Logistic regression, k-nearest neighbors, naïve Bayes classifier and random forests | Vodencarevic et al.[49] (Feb. 2021) | ||
| ML methods applied to genetic data in RA have demonstrated the potential to predict drug response, uncover molecular mechanisms of | Genetic data | GPR | Guan et al.[50] (Dec. 2019) |
| therapy, identify genetic markers associated | Text mining | Wang et al.[51] (Jul. 2021) | |
| with clinical treatment response outcomes, to specific and treatments. accurately predict | Random forest, SVM | Kim et al.[52] (Oct. 2021) | |
| Random forest | Tao et al.[53] (Oct. 2021) | ||
| Multivariate logistic regression, elastic net, random forest, and SVM | Kim et al.[54] (Oct. 2022) | ||
| Random forest | Lim et al.[55] (Oct. 2022) | ||
| Random forest | Lim et al.[56] (Jan. 2022) | ||
| Random forest | Myasoedova et al.[40] (Jun. 2022) |
DL, deep learning; NLP, natural language processing; SVM, support vector machine; GPR, Gaussian process regression; EHRs, electronic health records; RCTs, randomized controlled trials; XGBoost, eXtreme gradient boosting; COVID-19, coronavirus disease 2019; ML, machine learning.