From: Adapting transformer-based language models for heart disease detection and risk factors extraction
Risk factor | Time attribute | Precision | Recall | F-Score | Support |
---|---|---|---|---|---|
CAD | Before_DCT | 0.78 | 0.93 | 0.85 | 462 |
 | During_DCT | 0.00 | 0.00 | 0.00 | 94 |
 | After_DCT | 0.67 | 0.63 | 0.65 | 116 |
Diabetes | During_DCT | 0.49 | 0.33 | 0.39 | 134 |
 | Before_DCT | 0.78 | 0.85 | 0.81 | 557 |
 | After_DCT | 0.00 | 0.00 | 0.00 | 109 |
Hypertension | Before_DCT | 0.00 | 0.00 | 0.00 | 33 |
 | During_DCT | 0.79 | 0.87 | 0.83 | 363 |
 | After_DCT | 0.89 | 0.79 | 0.84 | 364 |
Obese | Before_DCT | 0.00 | 0.00 | 0.00 | 35 |
 | During_DCT | 0.89 | 0.75 | 0.82 | 65 |
 | After_DCT | 0.73 | 0.67 | 0.70 | 36 |
Medication | Before_DCT | 0.62 | 0.42 | 0.50 | 910 |
 | During_DCT | 0.67 | 0.34 | 0.45 | 785 |
 | After_DCT | 0.61 | 0.26 | 0.36 | 715 |
Hyperlipidemia | Before_DCT | 0.00 | 0.00 | 0.00 | 85 |
 | During_DCT | 0.66 | 0.95 | 0.78 | 160 |
 | After_DCT | 0.00 | 0.00 | 0.00 | 34 |
Weighted average | Â | 0.9251 | 0.9373 | 0.9284 | 42946 |
Macro average | Â | 0.3144 | 0.2846 | 0.2736 | 42946 |
Micro average | Â | Â | Â | 0.9373 | 42946 |
Accuracy | Â | Â | Â | 0.9373 | 42946 |