Machine learning to diagnose PCD

  • Research type

    Research Study

  • Full title

    Improving Primary Ciliary Dyskinesia diagnosis using artificial intelligence

  • IRAS ID

    259067

  • Contact name

    Claire Hogg

  • Contact email

    Hogg@rbht.nhs.uk

  • Sponsor organisation

    Royal Brompton and Harefield NHS Foundation Trust

  • Duration of Study in the UK

    2 years, 0 months, 0 days

  • Research summary

    In the airway, cells are lined with many hair-like cilia that sweep in a coordinated way to clear mucus, bacteria and debris from the airway. In primary ciliary dyskinesia (PCD), problems with the movement of the cilia result in mucus build-up that can lead to recurring infections and damage to the lungs. Around 1 in 10,000 people have PCD in the UK, but it’s much more common in some communities. Early diagnosis is important to help prevent lung damage in childhood and to keep the lungs working as well as they can. Diagnosis is complex and currently relies on looking at cilia (taken from the nose) and checking the ciliary ultrastructure with a powerful electron microscope. It can be lengthy and difficult to make a diagnosis and experts can have different opinions on what the electron micrographs show. The aim of this project is to use Artificial Intelligence computing techniques to identify both healthy and abnormal cilia from electron microscopy images to provide a faster, more accurate and reliable approach to evaluate ciliary structure.

  • REC name

    South Central - Oxford B Research Ethics Committee

  • REC reference

    19/SC/0229

  • Date of REC Opinion

    28 May 2019

  • REC opinion

    Further Information Favourable Opinion