Goals | Year of publication | Author | Journal | Input data | Machine learning technology | Predictive effectiveness |
Precision medicine | ||||||
Early screening and diagnosis | 20207 | Deodhar et al | Clin Rheumatol | Medical claims database medical records | Optimise the linear regression model A/B | Model A/B PPV: 6.24% Simplified linear regression model PPV: 2.55% |
20229 | Zhu et al | Rheumatol Ther | Laboratory indicators | LASSO, SVM-RFE, RF and Nomograms | Training cohort AUC: 0.87 Validation cohort AUC: 0.82 | |
202210 | Ye et al | Rheumatology | MRI and medical records | mRMR, LASSO, multivariable logistic regression analysis and Nomograms | Rad-score:training/validation cohort AUC: 0.82. The clinical-radiomics nomogram model: training/validation cohort AUC: 0.9 | |
202211 | Wen et al | Front Genet | GEO database | LASSO, SVM-RFE, RF and Nomograms | AUC>0.84 | |
202212 | Han et al | Front Immunol | GEO database | WGCNA, SVM-RFE | AUC>0.7 | |
20208 | Zhao et al | Rheumatology | Electronic health records | Logistic regression, LASSO penalised and multimodal automated phenotyping | AUC: 0.93 | |
Assisted classification and stratification | 202314 | Zhang et al | Journal of Digital Imaging | CT images | 3D convolutional neural network | Validation set AUC: 0.91, 0.80, 0.96 Test set AUC: 0.94, 0.82, 0.93 |
202015 | Castro-Zunti et al | Comput Med Imaging Graph | CT and age | RF, K-NN | AUC>0.9 | |
202216 | Bressem et al | Radiology | MRI | Deep learning | Inflammatory changes: AUC: 0.94 Structural changes: AUC: 0.89 | |
202217 | Lin et al | Rheumatology | MRI | Attention U-net | AUC: 0.92 | |
202018 | Tenório et al | Int J Comput Assist Radiol Surg | MRI-based radiomics biomarkers | Mann-Whitney U | AUC>0.8 | |
Monitoring the condition | 202320 | Baek et al | Arthritis Res Ther | mSASSS and clinical data | Artificial neural network, generalised linear model | ANN RMSE: 2.83 GLM RMSE: 2.99 |
201921 | Gossec et al | Arthritis Care Res (Hoboken) | Wearable activity trackers | Multiclass Bayesian | Positive predictive value 91% | |
Evaluate the efficacy of the drug | 202022 | Lee et al | Sci Rep | The baseline demographic and laboratory data | ANN,logistic regression, support vector machine, random forest and XGBoost models | ANN model AUC: 0.783, logistic regression, support vector machine, random forest and XGBoost models (AUC: 0.719, 0.699, 0.761 and 0.713) |
202223 | Wang et al | JAMA Netw Open | The baseline demographic, laboratory data and medication history | Logistic regression, linear discriminant analysis, SVM, GBM and RF | AUC>0.7 | |
202124 | Barata et al | JMIR Med Inform | Electronic medical records | Joint models | / | |
Prediction of comorbidities | 202227 | Zhang et al | Front Genet | GEO database | LASSO, RF, XGBoost and SVM | SVM: AS AUC 0.7, LBMD AUC: 0.76 |
202029 | Navarini et al | Rheumatol Ther | Clinical data | SVM, RF and KNN | AUC values for the ML algorithms were: 0.70 for SVM, 0.73 for RF and 0.64 for KNN |
ANN, artificial neural network; AUC, area under curve; 3D, three-dimensional; GBM, gradient boosting machine; GEO, Gene Expression Omnibus; GLM, generalised linear model; K-NN, K-nearest neighbour; LBMD, low bone density; PPV, positive predictive value; RF, random forest; RMSE, root mean square error; SVM-RFE, support vector machine recursive feature elimination; WGCNA, Weighted Gene Co-expression Network Analysis; XGBoost, extreme gradient boosting.