Clinical Deidentification Pipeline (Sentence Wise)

Description

This pipeline can be used to deidentify PHI information from medical texts. The PHI information will be masked and obfuscated in the resulting text. The pipeline can mask and obfuscate MEDICALRECORD, ORGANIZATION, PROFESSION, HEALTHPLAN, NAME, LOCATION-OTHER, URL, DEVICE, CITY, DATE, ZIP, STATE, COUNTRY, STREET, PHONE, LOCATION, EMAIL, IDNUM, BIOID, FAX, AGE, LOCATION_OTHER, DLN, CONTACT, ID, SSN, ACCOUNT, PLATE, VIN, LICENSE, IP entities.

Predicted Entities

ACCOUNT, AGE, BIOID, CITY, CONTACT, COUNTRY, City, Country, DATE, DEVICE, DLN, EMAIL, FAX, HEALTHPLAN, HOSPITAL, ID, IDNUM, IP, LICENSE, LOCATION, LOCATION-OTHER, LOCATION_OTHER, MEDICALRECORD, NAME, ORGANIZATION, PHONE, PLATE, PROFESSION, SSN, STATE, STREET, URL, VIN, ZIP

Live Demo Open in Colab Download Copy S3 URI

How to use

from sparknlp.pretrained import PretrainedPipeline

deid_pipeline = PretrainedPipeline("clinical_deidentification_nameAugmented_v2", "en", "clinical/models")

text = """Dr. John Lee, from Royal Medical Clinic in Chicago,  attended to the patient on 11/05/2024. 
The patient’s medical record number is 56467890. 
The patient, Emma Wilson, is 50 years old,  her Contact number: 444-456-7890 ."""

deid_result = deid_pipeline.fullAnnotate(text)

print(''.join([i.metadata['masked'] for i in deid_result['obfuscated']]))
print(''.join([i.result for i in deid_result['obfuscated']]))

import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline

val deid_pipeline = PretrainedPipeline("clinical_deidentification_nameAugmented_v2", "en", "clinical/models")

val text = """Dr. John Lee, from Royal Medical Clinic in Chicago,  attended to the patient on 11/05/2024. 
The patient’s medical record number is 56467890. 
The patient, Emma Wilson, is 50 years old,  her Contact number: 444-456-7890 ."""

val deid_result = deid_pipeline.fullAnnotate(text)

println(deid_result("obfuscated").map(_("metadata")("masked").toString).mkString(""))
println(deid_result("obfuscated").map(_("result").toString).mkString(""))

Results

Masked with entity labels
------------------------------
Dr. <NAME>, from <LOCATION> in <CITY>,  attended to the patient on <DATE>.
The patient’s medical record number is <MEDICALRECORD>.
The patient, <NAME>, is <AGE> years old,  her Contact number: <PHONE> .

Obfuscated
------------------------------
Dr. Rhodia Cera, from 252 Mchenry St in UNTERLAND,  attended to the patient on 18/06/2024.
The patient’s medical record number is 16109604.
The patient, Eulice Hickory, is 44 years old,  her Contact number: 540-981-1914 .

Model Information

Included Models

• DocumentAssembler
• SentenceDetectorDLModel
• TokenizerModel
• WordEmbeddingsModel
• NerDLModel
• NerConverterInternalModel
• WordEmbeddingsModel
• MedicalNerModel
• NerConverterInternalModel
• MedicalNerModel
• NerConverterInternalModel
• MedicalNerModel
• NerConverterInternalModel
• ChunkMergeModel
• ContextualParserModel
• ContextualParserModel
• ContextualParserModel
• ContextualParserModel
• ContextualParserModel
• ContextualParserModel
• ContextualParserModel
• RegexMatcherInternalModel
• ContextualParserModel
• ContextualParserModel
• TextMatcherInternalModel
• TextMatcherInternalModel
• TextMatcherInternalModel
• ContextualParserModel
• RegexMatcherInternalModel
• RegexMatcherInternalModel
• RegexMatcherInternalModel
• ChunkMergeModel
• ChunkMergeModel
• DeIdentificationModel
• Finisher