A multi-centre, prospective, double-blinded, pilot study evaluating artificial intelligence driven automatic detection of proximal deep vein thrombosis (DVT).

Deep vein thrombosis (DVT) is a condition where blood clots form in the veins of the body, most frequently the leg. It is a serious condition and if left untreated clots can move to the lungs, which can be fatal. Early diagnosis is vital. Normally three steps are needed for a DVT diagnosis: a GP appointment; a hospital DVT clinic appointment and a leg ultrasound scan. This process should take only 4 hours, but it often takes longer and therefore a precautionary anticoagulant (blood-thinning) drug is given. Nearly 90% of people investigated for a DVT have no clot. We have a new way of diagnosing DVT using a novel computer software technology (AutoDVT). This guides a non-specialist nurse to diagnose a DVT. Early diagnosis will eliminate unnecessary blood-thinning drugs, reduce healthcare costs, improve patient safety and satisfaction. We plan to test how accurate this new method is in this study.