Data Scientist in the Coronary Research Group

About the role

The postholder will contribute to development of our portfolio of clinical and health informatics platforms and research which sits within the King’s BHF Centre of Research Excellence:

(https://www.kcl.ac.uk/scms/bhf-centre)

and specifically work within the Coronary Research Group at KCL.  

The Coronary Research Group carries out translational research into the mechanisms and consequences of cardiac ischaemia with a view to developing and evaluating novel and personalised treatments for ischaemic heart disease. The group employs diverse research methodologies to achieve these aims, from exploration of the pathophysiological basis of disease states by detailed characterisation of patients during diagnostic and therapeutic procedures through designing and conducting multicentre randomised clinical trials to applying machine learning and Artificial Intelligence techniques to data extracted from routine health records. The work will also include areas relevant to the BHF Centre Themes including heart failure, obesity and metabolic diseases with the scope of extending the timeframe of the contract based on the receipt of renewed funding. The Group’s research is funded by peer-reviewed grants from charities (including the Academy of Medical Sciences, British Heart Foundation and Medical Research Council) and public funding (UK National Institute for Health Research and US National Institute for Health) as well as investigator-initiated commercial funding.

The primary purpose of this role is to develop a cohort identification tool for cardiovascular research using multimodal electronic health records data. This will require the development of both the tool itself and a number of clinical phenotyping algorithms. This new tool will be part of the CogStack ecosystem, currently being deployed to multiple hospitals for near real-time analytics for direct patient care, streamlining services and research

https://www.youtube.com/watch?v=XU9ls9J9AmM; https://doi.org/10.1101/123299;

https://www.youtube.com/watch?v=XU9ls9J9AmM; 

https://doi.org/10.1101/123299; https://cogstack.github.io/

A key component of this system is our clinical NLP tool MedCAT:

(https://arxiv.org/abs/1912.10166; https://github.com/CogStack/MedCAT). 

The role will build clinical NLP and phenotyping using electronic health records, putting these tools in the hands of clinical users. As such this tool is a foundational piece of work to enable a multitude of subsequent projects and grant applications which use this software to identify a cohort of patients from the EHR. 

This is a 50% FTE post (17.5 hours per week), and you will be offered a fixed term contract for 8 months.

About You

To be successful in this role, we are looking for candidates to have the following skills and experience:

Essential criteria

• PhD or equivalent experience in relevant area
• Knowledge or interest in Natural language processing ideally over big biomedical data
• Programming (one or more of the following: Python, R, Javascript etc)
• Software development experiences in large-scale academic projects and/or industry environment
• Data management in large scale and heterogeneous data spaces
• Software development
• Clinical Informatics

Desirable criteria

• Biomedical ontologies / Semantic Web technologies
• Information retrieval, data analytics and/or text mining on Electronic Health Records and biomedical literature
• Knowledge of machine learning / deep learning with an interest in the application to electronic patient records
• Understanding or interest in graph data models / Knowledge graph techniques
• Some statistical skills

Advert information