Xray of human body with unreadble text overlaid

Training in methodological skills for Prediction Modelling

Clinical prediction models can help healthcare professionals and patients make clinical decisions with the aim of improving patient outcomes and quality of care. Developing and implementing an accurate, generalisable, and robust prediction model needs - besides clinical expertise, clinicians, patients and other stakeholders’ involvement - modern statistical and data science skills.

Virtual Data training Centre

Funded by UK Research and Innovation, the NIHR Maudsley BRC prediction modelling supports the development of an innovative, interactive online learning program to provide a comprehensive and integrated understanding of the requirements of modern data science in health research in the 21st century. This virtual online training centre will offer flexible online courses in R and Phyton programming, Prediction modelling, machine learning/AI and Natural Language processing. The online modules will be hosted by the NIHR Maudsley BRC CoqStack servers. Modules will introduce and train to use a variety of large data analysis techniques, allowing researchers to process health record data for both research and real-world applications. The first modules will be launched in the summer of 2022. More information about the centre which is part of Kings Innovation Scholars programme: Big Data skills Training” can be found at Innovation Scholars: Big Data Training (Pillar 1) and here: Innovation Scholars Training

Applied Statistical Modelling and Health Informatics

Members of the BRC Prediction modelling group are also supporting the MSc, PG Cert, PG Dip “Applied Statistical Modelling and Health Informatics”, which provides training in core applied statistical methodology, machine learning and computational methodology necessary for the successful development of clinical prediction models. More information about the MSc can be found here: https://www.kcl.ac.uk/study/postgraduate-taught/courses/applied-statistical-modelling-health-informatics