23 June 2026

Amanda Kiemes, a senior data analyst with the AI Centre for Value Based Healthcare at Guy’s and St Thomas’ NHS Foundation Trust (FT), details how funding from the KHP Centre for Translational Medicine is helping the team support researchers with their data analysis.

Please describe your day-to-day work and research interest(s)?

As part of the Natural Language Processing (NLP) team of the AI Centre for Value Based Healthcare, we develop bespoke AI models for projects where data needs to be surfaced from Electronic Health Records (EHR).

The day-to-day ranges from scoping and translating healthcare questions into data needs, investigating data queries, and transforming unstructured free-text into structured data by training and refining AI models.

The research domain of our projects is determined by our collaborators who approach us for our expertise in natural language processing, so we get to learn and collaborate in all different medical domains.

What is the potential of your research to have a positive impact on patient care?

As we work on a variety of projects, the outcomes can range from direct patient care such as identification of patients, to improving clinical pathways through auditing existing processes, to long-term impact through aiding clinical research. 

You received funding from the KHP Centre for Translational Medicine (CTM), how has this helped your research?

Funding from the CTM enables the work of the NLP team, allowing us to support researchers with data access, governance, and analysis. Additionally, funding has enabled the transformation of KHP hospital Epic data into the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM), the gold-standard for transferable and standardised data structures to facilitate efficient data querying for research.

What are the next steps for your research and how would you like to see it develop?

GSTT hospital data infrastructure is a changing landscape and it is an exciting time for our work. New data access routes through the implemented OMOP CDM and the new data platform Snowflake means we can work on projects more efficiently, but also have access to richer data, meaning we can answer data needs with richer and larger datasets.

Our future work involves the training and use of newer AI models that address domain-wide issues and are usable for a wider range of projects.