Table 2.

Summary statistics of annotated datasets of Chinese discharge summaries and admission notes

Dataset	Notes	Type	Sentences	Characters	NER tasks
					Problems	Procedures	Tests	Medications	Total
Training	266	Admission	20 506	277 701	16 253	1500	7414	840	26 007
		Discharge	15 140	243 069	13 308	2995	8093	1757	26 153
		All	35 646	520 770	29 561	4495	15 507	2597	52 160
Test	134	Admission	10 287	139 885	8180	671	3754	361	12 966
		Discharge	7698	125 335	6851	1522	4021	787	13 181
		All	17 985	265 220	15 031	2193	7775	1148	26 147
Total	400	Admission	30 793	417 586	24 433	2171	11 168	1201	38 973
		Discharge	22 838	368 404	20 159	4517	12 114	2544	39 334
		All	53 631	785 990	44 592	6688	23 282	3745	78 307

NER, named entity recognition.