
Summary of studies using AI technology with novel observational data
Author | Sample size | Mean age | Date type | Method | AUC (%) | Sensitivity (%) | Specificity (%) | Accuracy (%) |
---|---|---|---|---|---|---|---|---|
Tariq et al. [48] | 116 (ASD) | 4.10yrs (ASD) | Behavioral features | ADTree, SVM, LR, RK, LİSVM Sparse 5 feature LR classifier | 89-92 | 90-100 | 1.13-100 | 94-100 |
Liu et al. [39] | 29(ASD) | 7.9 yrs (ASD) | Eye-tracking | SVM | 89.63 | 93.1 | 86.21 | 88.51 |
Li et al. [43] | 14(ASD) | 32 yrs (ASD) | Hand movement | NB, SVM, RF, DT | - | 57.1-85.7 | 68.8-87.5 | 66.7-86.7 |
Anzulewicz et al. [44] | 37(ASD) | 4.5 yrs (ASD) | Hand movement | RF, RGF | 88.1-93.2 | 76-83 | 67-88 | - |
Crippa et al. [45] | 15(ASD) | 3.5yrs (ASD) | Upper-limb movement | SVM | - | 82.2-100 | 89.1-93.8 | 84.9-96.7 |
ADTree: alternating decision tree, AI: artificial intelligence, ASD: autism spectrum disorder, AUC: area under the curve, DT: decision tree, LİSVM: linear support vector machine, LR: logistic regression, NB: naïve Bayes, RF: random forest, RGF: regularized greedy forest, RK: radial kernel, SVM: support vector machine, TD: typically developing, yrs: years