Multimodal

We link different types of data … safely and fairly

We link structured data, unstructured data (e.g. clinic notes), and imaging data (e.g. MRI, CT scan, X-ray, ultrasound) to build a more complete picture of patient health.

This means we can securely combine clinical notes, laboratory results and medical scans to help researchers and clinicians uncover deeper insights that improve care.

We are also expanding to include digital pathology and genomic data, adding more dimensions to our multimodal data platform. Patients can choose how their data is used, and we only use data with strict safeguards in place.

  • Structured - OMOP

    OMOP Extraction System

    By default, UCLH clinical data is provided for research in a format called the OMOP Common Data Model (CDM), also referred to as simply ‘OMOP’.

    OMOP-ES is our OMOP data Extraction Pipeline. It allows a diverse range of data and systems to be mapped to OMOP in a controlled, governed manner – with rules to anonymise and redact data so as to respect the privacy of patients.

    OMOP-ES easily allows new types of data to be incorporated, enforces high quality mapping and portability to other EPIC sites.

    Read more about our OMOP Policy on our Policy Page.

  • Unstructured - Cogstack

    We use Cogstack to anonymise free text data from clinical notes and imaging reports. A detailed description is available on our wiki.

    Read about our Cogstack anonymisation policy on our Policy page.

  • Imaging - PIXL

    PIXL Image eXtraction Laboratory (PIXL) is a system for extracting, linking and de-identifying DICOM imaging data, structured EHR data and free-text data from radiology reports at UCLH.

    Learn more about our PIXL anonymisation policy on our Policy page.

Interested in data?

Find out how to request data for research here.