Healthcare NLP Models

Learn about the Language service Healthcare NLP models to extract entities from healthcare records such as electronic health records (EHR), progress notes, and clinical trial documents.

The Healthcare NLP model is used to process healthcare text records such as EHR to extract entities, determine assertion statuses, identify related entities, and link those entities with SNOMED, ICD10 ontologies.

Healthcare NLP is a suite of four models:

Health Named Entity Recognition: Identifies key entities from text. This includes identifying medical conditions, medications, dosages, symptoms, test results, treatments, and procedures.
Relationship Extraction: Identify relationships between different medical entities. For example, it extracts the relationship between the medication and dosage.
Assertion Detection: Identifies contextual modifiers to the extracted entities such as the status, subject, time, and so on.
Medical Entity Linking: Link the extracted entities with codes from the biomedical vocabularies, SNOMED-CT and ICD-10.

Note

When working with the Oracle NLP model, it's important to review the provided confidence scores for accuracy. These scores can help you determine the appropriate confidence threshold for your particular use case. However, to ensure compliance with regulations, it's always advisable to verify the accuracy of any detected Health entities through other means such as human review.

Sample Use Cases

Healthcare NLP models have a wide range of use cases in healthcare, revolutionizing the industry by improving patient care, streamlining operations, and facilitating research.

Clinical Documentation Improvement: NLP models can analyze clinical notes and suggest improvements, ensuring accurate and complete patient records for billing and compliance purposes.
Clinical Decision Support: NLP can help providers by extracting relevant information from patient records to provide recommendations for treatment options.
Medical Coding: NLP can help automate the coding of medical procedures and diagnoses by analyzing physician notes.
Tele Medicine: Develop voice-activated assistants that can transcribe doctor-patient interactions, update electronic health records, and provide quick access to relevant patient data during appointments.

Supported Language

The PHI model supports English (United States).

Supported Entity Types


Entity Type	Description
`HEADER` `SUB_HEADER`	Document Section
`ANATOMICAL_SITE`	Anatomy
`SIGN_SYMPTOM` `DIAGNOSIS` `OBSERVATION` `MODIFIER`	Medical Condition
`MEDICINE_NAME` `MEDICINE_FREQUENCY` `MEDICINE_DOSE` `MEDICINE_DOSE.FORM` `MEDICINE_ROUTE` `MEDICINE_DURATION` `MEDICINE_STRENGTH` `MEDICATION_CLASS` `MEDICINE_REFILL_AMOUNT`	Medicine
`VITALS` `OBSERVABLE_ENTITY`	Physical Examination
`LAB_TEST`	Laboratory Examination
`TREATMENT` `PROCEDURE`	Treatment and Procedures
`MEASUREMENT`	Measurements
`DIRECTION` `RESULT`	General
`ALLERGEN_AGENT`	Allergy
`IMMUNIZATION`	Immunization
`ROLE`	ROLE
`FAMILY_RELATION`	FAMILY Relation

Supported Ontologies

SNOMED CT US
ICD-10-CM

Supported Assertion Types


Assertion Type	Possible Values
Status	Negated
Certainty	Certain Possible Hypothetical Conditional Uncertain
Temporality	Past Present Future
Subject	Physician Patient Family Other
Action	Start Stop Increase Decrease OtherChange
Severity	Mild Moderate Severe
Chronicity	Acute Chronic Acute-on-Chronic Subacute Major
Course	Worsening Improving Controlled Uncontrolled Other

Supported Relation Types


Relation Type	Subject Entity	Object Entity
`ANATOMICAL_SITE_OF_DIAGNOSIS`	`DIAGNOSIS`	`ANATOMICAL_SITE`
`ANATOMICAL_SITE_OF_OBSERVATIONS`	`OBSERVATION`	`ANATOMICAL_SITE`
`ANATOMICAL_SITE_OF_SIGN_SYMPTOM`	`SIGN_SYMPTOM`	`ANATOMICAL_SITE`
`ANATOMICAL_SITE_OF_PROCEDURE`	`PROCEDURE`	`ANATOMICAL_SITE`
`DOSAGE_OF_MEDICINE`	`MEDICINE_NAME`	`MEDICINE_DOSAGE`
`DURATION_OF_MEDICINE`	`MEDICINE_NAME`	`MEDICINE_DURATION`
`ROUTE_MODE_OF_MEDICINE`	`MEDICINE_NAME`	`MEDICINE_ROUTE_MODE`
`STRENGTH_OF_MEDICINE`	`MEDICINE_NAME`	`MEDICINE_STRENGTH`
`FREQUENCY_OF_MEDICINE`	`MEDICINE_NAME`	`MEDICINE_FREQUENCY`
`RESULT_OF_LAB_TEST`	`LAB_TEST`	`RESULT`
`MEASUREMENT_OF_LAB_TEST`	`PHYSICAL_EXAMINATION`	`MEASUREMENT`
`MEASUREMENT_OF_PHYSICAL_EXAMINATION`	`PHYSICAL_EXAMINATION`	`MEASUREMENT`
`RESULT_OF_VITALS`	`PHYSICAL_EXAMINATION`	`RESULT`
`RESULT_OF_OBSERVABLE_ENTITY`	`OBSERVABLE_ENTITY`	`RESULT`
`DIRECTION_OF_ANATOMICAL_SITE`	`ANATOMICAL_SITE`	`DIRECTION`
`ANATOMICAL_SITE_OF_TREATMENT`	`TREATMENT`	`ANATOMICAL_SITE`
`MODIFIER_OF_DIAGNOSIS`	`DIAGNOSIS`	`MODIFIER`
`MODIFIER_OF_PHYSICAL-EXAMINATION`	`PHYSICAL_EXAMINATION`	`MODIFIER`
`MODIFIER_OF_SIGN_SYMPTOM`	`SIGN_SYMPTOM`	`MODIFIER`
`MODIFIER_OF_OBSERVATION`	`OBSERVATION`	`MODIFIER`
`MODIFIER_OF_MEDICINE_NAME`	`MEDICINE_NAME`	`MODIFIER`
`MODIFIER_OF_MEDICINE_DOSAGE`	`MEDICINE_DOSAGE`	`MODIFIER`
`MODIFIER_OF_ANATOMICAL_SITE`	`ANATOMICAL_SITE`	`MODIFIER`
`TIME_OF_PROCEDURE`	`PROCEDURE`	`DATETIME`
`TIME_OF_SIGN_SYMPTOM`	`SIGN_SYMPTOM`	`DATETIME`
`TIME_OF_DIAGNOSIS`	`DIAGNOSIS`	`DATETIME`
`TIME_OF_LAB_TEST`	`LAB_TEST`	`DATETIME`