automatic_NLP_processor: Process NLP Annotations on the Current Patient Cohort

In simon-hans/CEDARS: Simple and Efficient Pipeline for Electronic Health Record Annotation

Process NLP Annotations on the Current Patient Cohort

Description

Accepts a list of patient ID's or alternatively can perform NLP annotations on all available patients in the database.

Usage

automatic_NLP_processor(
  patient_vect = NA,
  text_format = "latin1",
  nlp_engine = "udpipe",
  uri_fun = mongo_uri_standard,
  user,
  password,
  host,
  replica_set,
  port,
  database,
  max_n_grams_length = 7,
  negex_depth = 6,
  select_cores = NA,
  URL = NA
)

Arguments

patient_vect	Vector of patient ID's. Default is NA, in which case all available patient records will undergo NLP annotation.
text_format	Text format for NLP engine.
nlp_engine	Which NLP engine should be used? UDPipe is the only one supported for now.
uri_fun	Uniform resource identifier (URI) string generating function for MongoDB credentials.
user	MongoDB user name.
password	MongoDB user password.
host	MongoDB host server.
replica_set	MongoDB replica set, if indicated.
port	MongoDB port.
database	MongoDB database name.
max_n_grams_length	Maximum length of tokens for matching with UMLS concept unique identifiers (CUI's). Shorter values will result in faster processing. If 0 is chosen, UMLS CUI tags will not be provided.
negex_depth	Maximum distance between negation item and token to negate. Shorter distances will result in decreased sensitivity but increased specificity for negation.
select_cores	How many CPU cores should be used for parallel processing? Max allowed is total number of cores minus one. If 1 is entered, parallel processing will not be used.
URL	UDPipe model URL.

Value

Confirmation that requested operation was completed, or error message if attempt failed.

Examples

## Not run: 
automatic_NLP_processor(patient_vect = NA, text_format = 'latin1', nlp_engine = 'udpipe',
URL = 'models/english-ewt-ud-2.4-190531.udpipe', uri_fun = mongo_uri_standard, user = 'John',
password = 'db_password_1234', host = 'server1234', port = NA, database = 'TEST_PROJECT',
max_n_grams_length = 7, negex_depth = 6, select_cores = 1)

## End(Not run)

simon-hans/CEDARS documentation built on Feb. 14, 2024, 3:16 a.m.