SOS Study
Research type
Research Study
Full title
Sarcopaenia Opportunistic Screening for Frailty on Standard of Care CT (SOS study): Implementing a generalisable tool for sarcopaenia quantification in the NHS
IRAS ID
314668
Contact name
Andrea Rockall
Contact email
Sponsor organisation
Imperial College Healthcare NHS Trust
Duration of Study in the UK
1 years, 0 months, 1 days
Research summary
Supporting people to age well is one of three major categories of the NHS long term plan.
Sarcopaenia has emerged as a closely related independent predictor of adverse health outcomes across a spectrum of patient populations including cancer treatments, chemotherapy, emergency and elective surgery (4) falls, trauma (5) and liver disease (6-12). Sarcopaenia is correlated with physical disability, quality of life and life expectancy. It is recognised that exercise programmes and high protein diets can be used to rectify sarcopaenia in those patients found to have significant loss of muscle mass (13).
We have developed a vendor neutral AI tool (using a fully-convolutional-neural-network) that automatically identifies the third lumbar level (L3) and then segments the muscle, to automate quantification of muscle mass and composition.The innovation we propose allows fully automatic selection of the L3 level and fully automatic segmentation of the muscle. Currently, very few cases require manual corrections and thus the time taken for the output is minimal and indeed the ease of use would make this a tool that could fully translate the use of sarcopaenia measurements to clinical care in thousands of patients. The Sarcopenia tool works quickly (60 seconds) using a PC, though processing is faster with powerful processing unit (graphics processing unit, GPU), thus potentially cost effective in routine use.
REC name
East of England - Cambridge Central Research Ethics Committee
REC reference
22/EE/0101
Date of REC Opinion
23 May 2022
REC opinion
Favourable Opinion