IP6 - CHAIROS

  • Research type

    Research Study

  • Full title

    A study to assess the clinical and cost-effectiveness of the Galen Prostate Artificial Intelligence Histology system in diagnosing clinically important prostate cancer on prostate biopsy tissue slides

  • IRAS ID

    293128

  • Contact name

    Hashim Uddin Ahmed

  • Contact email

    hashim.ahmed@imperial.ac.uk

  • Sponsor organisation

    Imperial Research Governance and Integrity Team

  • Clinicaltrials.gov Identifier

    NCT05228197

  • Duration of Study in the UK

    2 years, 0 months, 1 days

  • Research summary

    Prostate biopsy tissue is processed in the laboratory and cut into smaller sections and placed on small glass slides for the pathology doctor to look at under a microscope. The purpose of this study is to carry out a very detailed testing of how well a new artificial intelligence (AI) software is in ruling-out and detecting prostate cancer on high resolution images taken of prostate biopsy tissue.

    The study is divided into two stages. The first is the Calibration Stage. This is where we approach patients who have already had prostate biopsies as part of their normal care in the past 24 months to kindly allow us to take high-resolution scans of their biopsy tissue slides. These high resolution images will be looked at by a pathologist who will draw out where all the different types of tissue are such as cancer, atrophy, inflammation, and high grade PIN in addition to any other features. These will then be used to fine-tune the AI software before we run the second Validation Stage.

    The Validation Stage will test how accurate the AI software is in 600 patients. Patients who are advised and agreed to have a prostate biopsy will be asked to consider participating. If they agree, and after the biopsies have been taken and processed by the NHS laboratory to create slides, we will take high resolution images of these slides. The AI software will evaluate these slides independently of the pathology doctor and vice versa. We will also ask independent pathologists to also look at the slides. The AI software will be compared to the pathology doctors' reports.

    If the AI software is shown to be accurate, then in future, we might be able to use the AI software in the NHS to help pathology doctors and patients.

  • REC name

    South West - Cornwall & Plymouth Research Ethics Committee

  • REC reference

    21/SW/0097

  • Date of REC Opinion

    29 Jul 2021

  • REC opinion

    Favourable Opinion