MAPS part 1

  • Research type

    Research Study

  • Full title

    Multi-parameter Anaesthetic Physiological Signals analysis (MAPS part 1 of 2)

  • IRAS ID

    160044

  • Contact name

    David Marshall

  • Contact email

    MarshallAD@cardiff.ac.uk

  • Duration of Study in the UK

    2 years, 0 months, 0 days

  • Research summary

    Heart failure is clinically significant because typically 2% of hospital patients are liable to it but only half are previously diagnosed before decompensating with consequent additional mortality and costs. Early management can therefore save both lives and costs to the NHS.

    We intend to explore if computational analysis of the arterial waveform in association with other discrete data, would yield features that may be useful as a future diagnostic tool for surgical patients who have not been previously diagnosed as having heart failure.

    Steps of study are: 1. Export and anonymise routinely monitored physiological waveform data in theatres and ITU from NHS servers; 2. Apply and refine computational algorithms originally developed on MIMIC (mimic.physionet.org) data to extract features from these waveforms; 3. Demonstrate the feature­ extraction algorithms can be used to compress a series of waveforms into exemplar waves without significant loss of information, therefore suitable for archiving.

    Comparing the sets of waveform features from different patient groups is part of future work.

  • REC name

    Wales REC 3

  • REC reference

    14/WA/1242

  • Date of REC Opinion

    16 Dec 2014

  • REC opinion

    Favourable Opinion