Feasibility of Automated Boundary Detection in Speckle Images

  • Research type

    Research Study

  • Full title

    Feasibility of Automated Boundary Detection and Line Identification in Speckle Images

  • IRAS ID

    241339

  • Contact name

    Moin Saleem

  • Contact email

    m.saleem@bristol.ac.uk

  • Duration of Study in the UK

    1 years, 10 months, 1 days

  • Research summary

    Fluid overload in children on dialysis is very difficult to assess accurately, and is clinically very important. Ultrasound detection of lung fluid is a very promising new method, and this is done by counting ‘B lines’ on lung ultrasound, which look like comet tails. Counting these by eye is operator dependent, and introduces unreliability.

    We propose that evaluation of B lines on lung ultrasound can be automated using signal processing techniques and analysis based on novel software techniques

    The objective of this study is to establish feasibility of the new technique in children on dialysis, and develop an automated protocol for ongoing use.

    Methods:
    10 haemodialysis patients aged 5 – 18 years will undergo a single 10 minute lung ultrasound examination at Clinical Research and Imaging Centre, Bristol, with Aplio 500 Ultrasound system.

  • REC name

    South West - Frenchay Research Ethics Committee

  • REC reference

    18/SW/0187

  • Date of REC Opinion

    31 Jan 2019

  • REC opinion

    Further Information Favourable Opinion