Artificial intelligence-augmented classifiers of lymph node pathology

  • Research type

    Research Study

  • Full title

    The development and validation of artificial intelligence-augmented classifiers of lymphoma and lymph node pathology.

  • IRAS ID

    257898

  • Contact name

    Christopher Carey

  • Contact email

    Christopher.Carey@newcastle.ac.uk

  • Sponsor organisation

    Newcastle upon Tyne Hospitals NHS Foundation Trust

  • Clinicaltrials.gov Identifier

    8946,

  • Duration of Study in the UK

    5 years, 0 months, 1 days

  • Research summary

    This is a study to develop and validate diagnostic classifiers of different types of lymphatic disorders, including malignant lymphoma, other cancers which have spread to lymph nodes, and non-cancerous inflammatory disorders. The study will use available diagnostic material and information, comprising high quality digital images of link-anonymised patient biopsies, and the original authorised biopsy report. Machine learning computer algorithms will learn how to identify and sub-classify these various disorders (so called 'artificial intelligence', or 'AI').

    The reason for doing this are various. The first is to improve diagnostic efficiency and capacity of the pathology laboratory. The incidence of cancer is steadily increasing, in line with an ageing population, and this has not been matched by an increase in the number of highly-trained pathologists. Using AI may also improve diagnostic efficiency, shortening the time from the biopsy to a final diagnosis, without compromising accuracy. Computers are able to work continuously, overnight and at weekends, and their assessments will be supervised by expert pathologists. We want to study to what extent this approach can be validated in the clinical setting, both as regards the diagnostic accuracy and to model potential health economic improvements compared to the current diagnostic workflow.

  • REC name

    East of England - Cambridgeshire and Hertfordshire Research Ethics Committee

  • REC reference

    19/EE/0285

  • Date of REC Opinion

    16 Oct 2019

  • REC opinion

    Further Information Favourable Opinion