AIM-Lung

  • Research type

    Research Study

  • Full title

    Can AI-based analysis of thoracic muscle and lung volumes on routine preoperative CT predict postoperative morbidity in cardiac surgery? A retrospective imaging sub-study of the Frail-Heart cohort.

  • IRAS ID

    368664

  • Contact name

    Mark J Bennett

  • Contact email

    mark.bennett2@wales.nhs.uk

  • Sponsor organisation

    Swansea Bay University Health Board

  • Duration of Study in the UK

    0 years, 5 months, 31 days

  • Research summary

    The Frail-Heart study was a retrospective observational cohort that explored the relationship between frailty and postoperative outcomes in cardiac surgery. It included a range of clinical, physiological, and functional assessments, and has recently been published. Within this cohort, 150 patients underwent routine preoperative chest CT imaging as part of a standardised protocol to rule-out Covid-19 immediately before elective and urgent cardiac surgery. These scans represent a valuable opportunity for retrospective image analysis to explore whether quantitative thoracic metrics can help predict postoperative morbidity.
    This sub-study (AIM-Lung) will use open-source imaging tools to extract thoracic skeletal muscle and lung features from these CT scans. Sarcopenia (loss of skeletal muscle mass and quality) and compromised lung function have both been implicated in adverse surgical outcomes. We hypothesise that CT-derived measurements may act as imaging biomarkers to identify high-risk patients.

  • REC name

    N/A

  • REC reference

    N/A