Synthetic healthcare/Synthetic

Patient records

Encounter-level records for healthcare model development and software testing. Diagnosis fields use ICD-10 codes. Identifiers are synthetic. This set does not contain real patients.

Volume: 750,000records
Formats: CSV · Parquet · JSONL
Label: Synthetic

Purchase

Sold on Datarade. This site does not take payment.

Placeholder — listing URL not set

Schema

Column preview for the launch specification. Names and types describe the file. They are not a sample of rows.

patient_idstring
Synthetic patient identifier. Not a real person.
birth_yearinteger
Year of birth.
sexstring
Recorded sex category.
encounter_datedate
Date of the encounter.
icd10_codestring
ICD-10 diagnosis code.
diagnosisstring
Short description paired with the ICD-10 code.
care_settingstring
Where the encounter happened, such as outpatient or inpatient.

Columns in Patient records
Column	Type	Description
patient_id	string	Synthetic patient identifier. Not a real person.
birth_year	integer	Year of birth.
sex	string	Recorded sex category.
encounter_date	date	Date of the encounter.
icd10_code	string	ICD-10 diagnosis code.
diagnosis	string	Short description paired with the ICD-10 code.
care_setting	string	Where the encounter happened, such as outpatient or inpatient.

Use

Develop and test software that reads coded clinical records
Train models that use ICD-10 diagnoses

Generation

Synthetic patient records. Diagnosis fields use ICD-10 codes. There are no real patients.

Every dataset is 100% synthetic. It is not copied from real customers, cardholders, or patients.
Each product is labeled as synthetic in the title, the description, and the file metadata.
Delivery includes a generation note: the methods and algorithms used, the schema, and known limitations.

Request a different schema