HDR UK Gateway
HDR Gateway logo

Bookmarks

King's College Hospital MedCAT NLP 2011-2019

Population Size

1,073,183

People

Population Size statistic card

Years

2011 - 2019

Years statistic card

Associated BioSamples

None/not available

Associated BioSamples statistic card

Geographic coverage

United Kingdom

England

Geographic coverage statistic card

Lead time

Not applicable

Lead time statistic card

Summary

SNOMED codes derived from free text EHR from 2011-2019 covering all inpatients at King's College Hospital, using MedCAT Natural Language Processing. Research use of the dataset is governed by the KERRI committee, and requires a KCH principal investigator.

Documentation

This dataset contains Natural Language Processing (NLP) output from the MedCAT library applied to the full text content of the King's College Hospital electronic health record available through CogStack. Documents were annotated with SNOMED codes and meta-annotations for experiencer, negation and temporality.

Research use of the dataset is governed by the patient-led KERRI committee, and requires a KCH principal investigator.

Dataset type

Health and disease

Dataset sub-type

Not applicable

Dataset population size

1073183

Associated media

https://khp-informatics.github.io/NLP-embedding-visualisation/

Keywords

Observations

Observed Node

Disambiguating Description

Measured Value

Measured Property

Observation Date

Persons

1073183

count

31 Dec 2019

Provenance

Purpose of dataset collection

Care

Collection source setting

Secondary care - In-patients

Image contrast

Not stated

Biological sample availability

None/not available

Structural Metadata

Details

Publishing frequency

Continuous

Version

1.0.0

Modified

08/10/2024

Citation Requirements

King's College London NHS Foundation Trust

Coverage

Start date

01/01/2011

End date

31/12/2019

Time lag

Not applicable

Geographic coverage

United Kingdom, England, London

Minimum age range

18

Maximum age range

100

Accessibility

Language

en

Controlled vocabulary

SNOMED CT

Format

text/json, text/csv

Data Access Request

Dataset pipeline status

Not available

Access rights

https://www.kch.nhs.uk/research/use-of-data-for-research

Time to dataset access

Not applicable

Access method category

Varies based on project

Access service description

Research use of the dataset is governed by the patient-led KERRI committee, and requires a KCH principal investigator. We recommend making contact with a KCH principal investigator first to facilitate applications for approvals. The data will only be accessible in the KCH data environment within the NHS firewall and will not be transferred out of KCH.

Jurisdiction

GB-ENG

Data use limitation

Research use only

Data use requirements

User-specific restriction,Project-specific restrictions

Data Controller

Kings College Hospital NHS Foundation Trust, with oversight of the Caldicott Guardian

Data Processor

N/A

Dataset Types: Health and disease

Collection Sources: Secondary care - In-patients