Keywords:
Artificial Intelligence, eHealth, Vascular, Ultrasound, Diagnostic procedure, Embolism / Thrombosis
Authors:
J. Oppenheimer, R. Mandegaran, B. Kainz, M. P. Heinrich, F. Noor, S. Mischkewitz, A. Ruttloff, P. Klein-Weigel
DOI:
10.26044/ecr2022/C-10357
Purpose
Deep vein thrombosis (DVT) is a blood clot of the deep veins, most commonly of the lower limbs, which can potentially lead to a deadly pulmonary embolism (PE). The diagnosis of a DVT is made in a multi-step process consisting of clinical scores, blood-tests, and the gold-standard compression ultrasound examination of the leg veins. This exam is usually performed by a highly trained radiologist or sonographer [1, 2]. Patients often occupy valuable resources and beds in emergency departments while waiting for a scan or must be transported from the ward to the ultrasound department, leading to delays. Scanning by a specialist is costly and prolongs time to diagnosis and subsequent treatment.
AutoDVT (ThinkSono GmbH, Potsdam, Germany), a novel machine-learning software, provides a tool to aid non-specialists in acquiring appropriate compression sequences for remote DVT assessment. Ultrasound clips can then be reviewed by an expert remotely to triage suspected DVT patients better, as well as potentially diagnose them. This could result in decreased costs due to better and earlier triaging and diagnosis of patients.