Can artificial intelligence pre-screen colonoscopy lesions

  • Research type

    Research Study

  • Full title

    Can artificial intelligence pre-screen colonoscopy lesions to improve bowel cancer screening

  • IRAS ID

    348024

  • Contact name

    Philip Quirke

  • Contact email

    p.quirke@leeds.ac.uk

  • Sponsor organisation

    University of Leeds

  • Duration of Study in the UK

    3 years, 0 months, 1 days

  • Research summary

    The bowel cancer screening programme investigates 2M people annually as well as routine colonoscopy samples being removed from symptomatic patients from the straight to FIT programme. This represents a significant burden on histopathology services. Artificial intelligence applied on Haematoxylin and Eosin sections may be able to prescreen lesions allowing some very low risk lesions not to be reviewed by Pathologists thus potentially saving medical resource. Better classification and grading of lesions may be possible to achieve using AI as well as improved reproducibility. We wish to develop an algorithm that is cost effective and of benefit to the NHS to allow rapid assessment of lesions removed at either screening colonoscopy or ‘straight to fit’ colonoscopy improving diagnoses, reducing the burden on Pathologists and freeing up time to increase capacity elsewhere in the system.

  • REC name

    North East - Newcastle & North Tyneside 2 Research Ethics Committee

  • REC reference

    24/NE/0232

  • Date of REC Opinion

    16 Jan 2025

  • REC opinion

    Further Information Favourable Opinion