What is DVT?
Deep vein thrombosis (DVT) is a blood clot that forms within the deep vein in the body, most commonly in the leg. This can be potentially fatal if parts of the clot breaks off, travels through the body and becomes lodged in the lungs, this is known as a pulmonary embolism (PE).
PE and DVT together is called venous thromboembolism (VTE), and is considered a major cause of morbidity and mortality worldwide.
The current DVT diagnostic pathway is slow and inefficient. The standard way to detect a DVT is through a compression ultrasound exam, usually carried out by an expert (sonographer or radiologist).
However patients typically have multiple appointments, including a blood test and long wait times before they see the expert. Additionally, more than 80% of patients come back negative when investigated for DVT. This results in higher diagnostic costs and lower patient outcomes.
ThinkSono AI Solution
AutoDVT aims to enable non-radiology staff to also detect DVT by automatically guiding them through a compression ultrasound exam. The whole exam takes between 5 and 15 minutes using only a handheld device and a mobile phone.
With ThinkSono AI, the diagnosis can be done at the point of care, within 15minutes, and by non-radiology trained staff. This results in a shorter clinical pathway, lower diagnostic cost and improved patient outcomes.
Our Scientific Publications
AutoDVT: Joint Real-time Classification for Vein Compressibility Analysis in Deep Vein Thrombosis Ultrasound Diagnostics.
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.
Non-invasive Diagnosis of Deep Vein Thrombosis from Ultrasound with Machine Learning.
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 in a sensitivity of 93.8% and a specificity of 84.2%, a positive predictive value of 90.9%, and a negative predictive value of 88.9% compared to the clinical gold standard.