Adapting transformer-based language models for heart disease detection and risk factors extraction

Journal of Big Data

Table 7 BioClinicalBERT Evaluation Metrics for each risk factor indicator

Risk Factor	Indicator	Precision	Recall	F1-score	Support
CAD	Mention	0.86	0.92	0.88	259
	Event	0.80	0.81	0.80	258
	Test	0.93	0.42	0.57	65
	Symptom	0.85	0.80	0.82	99
Diabetes	Mention	0.99	0.99	0.99	601
	A1c	0.75	0.92	0.83	66
	Glucose	1.00	0.38	0.55	40
Hypertension	Mention	1.00	0.99	1.00	602
	High bp	0.99	1.00	0.99	206
OBESE	Mention	0.96	1.00	0.98	117
	Obese_BMI	0.00	0.00	0.00	5
Hyperlipidemia	Mention	0.97	0.99	0.98	286
	High chol.	0.00	0.00	0.00	10
	High LDL	0.79	0.79	0.79	29
Smoker	Smoker_never	0.94	0.95	0.94	114
	Smoker_ever	0.00	0.00	0.00	3
	Smoker_current	0.00	0.00	0.00	40
	Smoker_past	0.88	0.77	0.82	122
	Smoker_unknown	0.99	0.99	0.99	202
Medication		0.88	0.74	0.80	7253
Family history		0.0000	0.0000	0.0000	13
Weighted average		0.9338	0.9403	0.9357	42946
Macro average		0.3715	0.3614	0.3505	42946
Micro average				0.9403	42946
Accuracy				0.9403	42946