Publications & News
IN PROGRESS: Q3 2022
ADVENT: A Multi-Site Study to Validate The Efficacy of ThinkSono AI-Guidance for Use By Non-Specialist Practitioners in the UK
Abstract: In Progress: A prospective clinical implementation of AutoDVT in hospital DVT patient assessment settings (under research license). Key endpoints under measurement include data collection pathway improvement, clinical pathway health economic assessment, and quality of data capture by non-specialist healthcare practitioners (e.g. nurses). Participating NHS trusts include Oxford, Leicester, Sheffield, King’s, Barts, Buckinghamshire, St. George’s, Surrey & Cardiff. Estimated completion: late 2022.
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTIONS – 2018
AutoDVT: Joint Real-time Classification for Vein Compressibility Analysis in Deep Vein Thrombosis Ultrasound Diagnostics.
Abstract: Abstract: We propose a dual-task convolutional neural network (CNN) to fully automate the real-time diagnosis of deep vein thrombosis (DVT). DVT can be reliably diagnosed through evaluation of vascular compressibility at anatomically defined landmarks in streams of ultrasound (US) images.
NATURE DIGITAL MEDICINE (NPJ) – 2021
Non-invasive Diagnosis of Deep Vein Thrombosis from Ultrasound with Machine Learning.
Abstract: Abstract: We train a deep learning algorithm on ultrasound videos from 246 healthy volunteers and evaluate on a sample size of 51 prospectively enrolled patients from an NHS DVT diagnostic clinic. 32 DVT-positive patients and 19 DVT-negative patients were included. Algorithmic DVT diagnosis results …
AI-guided novice-user compression sonography with remote expert DVT diagnosis
Abstract: 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.
Machine-learning software aids non-experts in performing “safe and efficient” remote DVT triage
Abstract: The machine-learning software was able to aid non-experts in acquiring valid ultrasound images of venous compressions and allowed safe and efficient remote triaging. Given that the vast majority of the requested DVT scans are negative, such a triaging strategy allows faster diagnosis and treatment of high-risk patients and can spare the need and cost of multiple unnecessary duplex scans. Patient waiting times can be reduced, and radiologist and sonographer resources can be reallocated.