COIOSDeep risk insights through multi-disciplinary research across the globe
a body of water with waves
Item new Preprint · not peer-reviewed Added

AI-guided ultrasound by operators without ultrasound training detected 93% of leg clots

Scans taken under AI guidance by operators without ultrasound training identified 92.9% of proximal deep vein thromboses, with 98.0% specificity in flagging who still needed a standard ultrasound.

A multicentre study of 594 patients compared AI-guided ultrasound by operators without ultrasound training against standard-of-care scans, with 67 DVTs across 700 scans. Clinician review of the AI-guided scans gave sensitivity of 92.9%, and 98.0% specificity in identifying patients still requiring a standard-of-care scan. Standard-of-care ultrasound was avoidable in 35.3% of patients, and 86.8% of AI-guided scans reached diagnostic image quality.

Why it is interesting: The AI changes who acquires the scan rather than who reads it, measured against concurrent standard care.

Source
medRxiv, 27 September 2026
Paper
Validating Artificial Intelligence Guidance for Ultrasound Acquisition and Remote Interpretation
Authors
Maldonado T, Muluk S, Rali P, Soni N, Nathanson R, Kuttab H et al.
DOI
10.64898/2026.07.16.26356882
Type
Preprint
Design
Multicentre, double-blinded, prospective non-randomised diagnostic accuracy study; 594 patients analysed, 700 scans, 67 proximal DVTs
Verdict
New finding
Driver
AI in medicine

Other drivers in Data, models and systems