From: Neural network methods for diagnosing patient conditions from cardiopulmonary exercise testing data
Model | Condition | Precision | Recall | F1 Score | Accuracy |
---|---|---|---|---|---|
flowchart (Hansen) | Heart Failure | 1.00 | 0.53 | 0.70 | 70 |
 | MetSyn | 0.76 | 0.87 | 0.81 |  |
flowchart (FRIEND) | Heart Failure | 0.78 | 0.93 | 0.85 | 77 |
 | MetSyn | 1.00 | 0.60 | 0.75 |  |
PCA + Logistic Regression | Heart Failure | 0.93 | 0.87 | 0.90 | 90 |
 | MetSyn | 0.88 | 0.93 | 0.90 |  |
AE +Logistic Regression | Heart Failure | 0.94 | 1.00 | 0.97 | 97 |
 | MetSyn | 1.00 | 0.93 | 0.97 |  |
CNN | Heart Failure | 1.00 | 0.80 | 0.86 | 90 |
 | MetSyn | 1.00 | 1.00 | 0.92 |  |