From: Machine learning models for identifying pre-frailty in community dwelling older adults
Models | Accuracy | Specificity | Sensitivity | Precision | F1-Score |
---|---|---|---|---|---|
Fried Frailty Phenotype Classification (not frail: prefrail) | |||||
 63 features | |||||
  Logistic Regression | 65.6 | 73.7 | 51.1 | 57.1 | 51.9 |
  Linear Discriminant Analysis | 65.1 | 78.9 | 53.2 | 57.9 | 53.4 |
  Support Vector Machine | 65.9 | 81.6 | 45.8 | 58.3 | 52.5 |
  Random Forest | 71.3 | 77.4 | 62.5 | 61.8 | 61.9 |
 Selected features | |||||
  Logistic Regression | 69.1 | 80.3 | 60.4 | 67.5 | 60.0 |
  Linear Discriminant Analysis | 69.1 | 88.0 | 60.4 | 75.0 | 60.0 |
  Support Vector Machine | 69.2 | 82.3 | 58.3 | 66.2 | 60.5 |
  Random Forest | 70.8 | 79.0 | 60.5 | 65.2 | 61.2 |
Clinical Frailty Scale Classification (not frail: prefrail) | |||||
 63 features | |||||
  Logistic Regression | 74.0 | 75.3 | 72.6 | 52.5 | 55.4 |
  Linear Discriminant Analysis | 73.8 | 76.4 | 58.8 | 52.8 | 56.6 |
  Support Vector Machine | 74.0 | 80.7 | 70.6 | 53.1 | 55.1 |
  Random Forest | 72.4 | 70.8 | 73.5 | 50.0 | 57.7 |
 Selected features | |||||
  Logistic Regression | 77.9 | 81.8 | 79.4 | 60.4 | 61.55 |
  Linear Discriminant Analysis | 77.1 | 77.5 | 73.5 | 56.2 | 62.36 |
  Support Vector Machine | 78.7 | 80.0 | 76.5 | 60.9 | 62.38 |
  Random Forest | 74.8 | 77.8 | 73.5 | 53.9 | 59.94 |