Table 1.

Summary of the mentions and CUIs in each semantic type group

		Full		Train		Test
Semantic type	Type description	Mention	CUI	Mention	CUI	Mention	CUI	Mention examples
DiseaseDisorder	Disease or Syndrome, Acquired Abnormality, etc.	2460	799	1242	471	1218	535	“congestive heart failure,” “coronary artery disease,” “diabetes mellitus”
Exam	Laboratory or Test Result, Laboratory Procedure, etc.	2263	463	1142	308	1121	318	“creatinine,” “hematocrit,” “platelet count”
Medication	Clinical Drug, Pharmacologic Substance, etc.	1816	522	959	353	857	345	“Coumadin,” “Lasix,” “Colace,” “Aspirin,” “Percocet”
Concept	Qualitative Concept, Spatial Concept, Quantitative Concept, etc.	1749	396	798	263	951	300	“some,” “left,” “mild,” “nontender,” “elevated,” “multiple”
Finding	Finding, Phenomenon or Process, etc.	1376	504	630	300	746	334	“afebrile,” “mass,” “No known drug allergies,” “lesion,” “weight loss”
Procedure	Therapeutic or Preventive Procedure	1138	385	573	226	565	253	“the procedure,” “chemotherapy,” “resection,” “radiation therapy”
SignSymptom	Sign or Symptom	992	183	535	128	457	127	“pain,” “nausea,” “chest pain,” “shortness of breath,” “fever”
Activity	Health Care Activity, Social Behavior, etc.	554	72	249	38	305	51	“blood pressure,” “heart rate,” “cardiac catheterization,” “pulse,” “evaluation”
Anatomy	Body Part, Organ, or Organ Component, Body Location or Region, etc.	530	294	248	160	282	192	“left lower lobe,” “bibasilar,” “LAD,” “lower extremity,” “transverse mesocolon”
Others	Bacterium, Body Substance, Medical Device, etc.	363	173	157	83	206	123	“drains,” “bacteria,” “yeast,” “protocol,” “somomabodies”
CUI-less	No corresponding CUI in UMLS	368	1	151	1	217	1	“no acute distress,” “further,” “apparent,” “atraumatic,” “workup”
Total		13 609	3791	6684	2330	6925	2578

CUI: concept unique identifier; UMLS: Unified Medical Language System.