Table 3.
The best feature combinations that yield the highest predictive accuracy of statistical models for heart toxicity‐based classification. Features that are consistently selected by all the models are bold
| Nearest neighbors | Linear SVM | RBF SVM | Decision Trees | Random Forest | Ada‐boost | Naive Bayes | Linear DA | Quadratic DA | |
|---|---|---|---|---|---|---|---|---|---|
| Selected Features | VolB | VolB | VolB | VolB | VolB | VolB | VolB | VolB | VolB |
| DisH‐B | Thickness | DisH‐B | DisH‐B | VolH | |||||
| VolipsL | |||||||||
| DisH‐B | |||||||||
| Laterality | |||||||||
| Thickness | |||||||||
| Accuracy | 0.83 | 0.77 | 0.83 | 0.73 | 0.77 | 0.73 | 0.83 | 0.83 | 0.93 |
VolB: Breast Volume; VolH: Heart Volume; VolipsL: IpsiLung Volume; DisH‐B: Distance between Breast and Heart; Thickness: Deep breath motion thickness variation.