Improving lesion recognition using routinely collected endoscopic and pathology data

To improve early detection of cancer and other diseases in gastroenterology, the quality of endoscopic investigations needs to be standardised and optimised. This requires effective training and continuous feedback on endoscopist performance.

Various performance metrics, such as adenoma detection rates (a type of colonic polyp that can lead to cancer), have been shown to directly reflect missed cancer rates for individual endoscopists. This metric, along with others, is derived from endoscopic and pathological datasets, which are currently manually analysed and fed back to endoscopists on a six-monthly basis. This is laborious, poorly executed and provides no feedback on how to improve endoscopic performance.

This Advancing Applied Analytics project aims to introduce automated feedback with suggestions for improvement in endoscopic performance from disparate datasets (pathology, and endoscopy with images), which will improve lesion recognition and therefore early detection. It will be implemented at the gastroenterology department at Guy’s and St Thomas’ NHS Foundation Trust, which performs around 20,000 endoscopies a year.

The project will involve developing a framework, using the programming language R, to allow data linkage between endoscopy images and reports, and pathology reports for individual endoscopists on a monthly basis. Automated data extraction, analysis and visualisation will be developed, as well as metrics to demonstrate deficiencies in lesion recognition.

The project will lay the foundations for an analytical infrastructure at the Trust, and will be used as an example of the benefits of using data analytics, and as a training tool to further expand the use of R within the Trust.

Contact information

For more information about this project, please contact Sebastian Zeki, Consultant Gastroenterologist, Guy’s and St Thomas’ NHS Foundation Trust.