Skip to main content

Table 3 Database entry data configuration

From: SemClinBr - a multi-institutional and multi-specialty semantically annotated corpus for Portuguese clinical NLP tasks

Field

Data type

Occurrence-id

Number

Patient-id

Number

Gender

Text

Birth date

Date

Inclusion date

Date

Discharge date

Date

Discharge type

Text

Discharge reason

Text

ICD-10

Text

Medical specialty

Text

Care reason

Text

Main complaint

Free-Text

History of disease

Free-Text

Past history

Free-Text

Family history

Free-Text

Physical examination

Free-Text

Main diagnosis hypothesis

Free-Text

Initial plan

Free-Text

Observations

Free-Text

  1. Data fields for each EHR entry in our main data source. The fields have different data types: numerical, date, text (one-line small text), and free-text (multi-line and large text)