Artificial Intelligence in diagnostic imaging of periodontal disease

  • Research type

    Research Study

  • Full title

    The application of artificial intelligence and deep learning models in the radiological diagnosis of dental disease

  • IRAS ID

    341577

  • Contact name

    Damian JJ Farnell

  • Contact email

    farnelld@cardiff.ac.uk

  • Sponsor organisation

    Cardiff University

  • Duration of Study in the UK

    3 years, months, days

  • Research summary

    This study is conducted in two phases to employ advanced dental imaging diagnostics, using Artificial Intelligence (AI) and deep learning (DL). In Phase 1, anonymized dental panoramic images are collected and analyzed. Phase 2 extends this work by including anonymized cone-beam computed tomography (CBCT) images. Periodontal pathology in the images is carefully labeled for training purposes. The data is secondary, sourced from existing dental records and anonymised at source.

    Various DL models will be applied to these labeled datasets to train and test their performance in diagnosing dental conditions. The best-performing models are identified, with their results saved for future use.

    The final stage of the research explores practical aspects of implementing any developed software, considering its benefits, challenges, and potential for real-world application. This includes addressing possible pitfalls (and evaluating the feasibility) of translating the research outcomes into clinical practice, paving the way for innovation in dental diagnostics in the NHS primary and secondary care settings.

  • REC name

    N/A

  • REC reference

    N/A