Deep MR-only RT

  • Research type

    Research Study

  • Full title

    Deep-Learning Magnetic Resonance-only Radiation Therapy

  • IRAS ID

    265421

  • Contact name

    Hazel McCallum

  • Contact email

    hazel.mccallum3@nhs.net

  • Sponsor organisation

    09382

  • Duration of Study in the UK

    1 years, 11 months, 31 days

  • Research summary

    Radiotherapy planning requires medical imaging information for accurately targeting the cancer cells and planning the treatment. Current practice is based on a complicated workflow requiring Magnetic Resonance (MR) and Computed Tomography (CT) imaging sessions, which is inconvenient for the patient and prone to errors in combining the images. The purpose of this study is to develop and evaluate a comprehensive MR-only radiotherapy workflow based on novel MR imaging and Deep Learning image processing. This will provide all the information required for accurate targeting of the cancer cells and developing the radiotherapy treatment plan. This will improve the accuracy of the treatment and the patient's experience by reducing the number of hospital visits required. The MR imaging protocol will also be designed to use silent MR imaging methods to improve the patient experience of the scan. Patient feedback on this aspect will be sought. This study will also explore the potential benefits of combined Positron Emission Tomography (PET)-MR images in planning radiotherapy treatments of pelvic cancers.

  • REC name

    London - Dulwich Research Ethics Committee

  • REC reference

    20/LO/0583

  • Date of REC Opinion

    27 May 2020

  • REC opinion

    Further Information Favourable Opinion