Table 3 Comparison with state-of-the-art ASCVD prediction using NHANES data
Researches | Purpose | Features | Sample size | Algorithms | AUC | Year of publication |
|---|---|---|---|---|---|---|
Chen et al. [22] | Identify chronic heart disease | 20 variables | 14,971 | support vector machine (SVM) | 0.898 | 2023 |
Inoue et al. [23] | Predict low HbA1c levels and all-cause or cardiovascular mortality | 72 variables | 39,453 | SuperLearner | 0.90 | 2020 |
Li et al. [24] | Heavy metals’ exposure to identify coronary heart disease | 13 heavy metals | 12,554 | random forest (RF) | 0.827 | 2023 |
Martin-Morales et al. [25] | Predict cardiovascular disease mortality | 59 variables | 9,706 | RF | 0.88 | 2023 |
Wang et al. [26] | Predict coronary heart disease risk in patients with periodontitis | 29 variables | 3,245 | K-nearest neighbor | 0.977 | 2023 |