An innovation funded by CW+ has demonstrated how artificial intelligence (AI) can help clinicians diagnose suspected blood clots faster, reducing waiting time and improving the experience for patients at Chelsea and Westminster Hospital.
Supported through CW Innovation, the three-month pilot tested a new AI-enabled ultrasound platform designed to support the assessment of patients with suspected deep vein thrombosis (DVT), a potentially serious condition caused by a blood clot that usually develops in a deep vein in the leg.
During the pilot, most patients received a clinical decision following their first appointment, with the typical time from starting a scan to receiving a clinical decision taking less than 30 minutes.
Faster diagnosis and fewer hospital visits
Developed by ThinkSono, the technology combines a handheld ultrasound device with real-time AI guidance. It enables trained clinicians to carry out a point-of-care ultrasound examination during a patient’s first appointment, reducing the need for patients to return for a sonographer-led scan.
The system guides clinicians through the scanning process, while ultrasound images are reviewed remotely by specialist sonographers or radiologists. This helps clinicians determine more quickly whether a patient requires further investigation or treatment.
During the evaluation, clinicians completed 120 examinations and ruled out DVT in 97 cases. The typical time from starting a scan to receiving a clinical decision was less than 30 minutes, with 80% of patients safely discharged following their first scan.
For many patients, this meant avoiding longer waits, repeat appointments and unnecessary treatment while awaiting a diagnosis.
Improving patient experience and reducing pressure on services
Fewer than 20% of patients required a follow-up departmental ultrasound after their ThinkSono scan. For those who did need further investigation, the average wait time from first presentation to diagnosis was reduced from 19 hours to nine hours, enabling patients diagnosed with DVT to receive their diagnosis on the same day.
Within four weeks of introducing the new AI-supported pathway, the number of patients receiving a scan report within the National Institute for Health and Care Excellence (NICE) recommended four-hour timeframe increased from 18% to 51%.
The evaluation also demonstrated how trained nurses, advanced clinical practitioners and resident doctors can undertake point-of-care DVT scanning using AI guidance, helping to address the national shortage of qualified sonographers while maintaining specialist oversight through remote review.
Supporting innovation that delivers impact
The pilot took place in Same Day Emergency Care (SDEC) at Chelsea and Westminster Hospital and was highlighted in the Trust’s recent Care Quality Commission (CQC) report as demonstrating a forward-looking approach to service delivery, with the potential to reduce unnecessary admissions and improve waiting times.
Dr Jonathan Downing, Consultant Acute Physician and Service Lead for Same Day Emergency Care at Chelsea and Westminster Hospital NHS Foundation Trust, said: ‘This is a fundamental change in how we manage suspected lower-limb DVT.
‘Instead of waiting hours, and sometimes overnight, for access to ultrasound imaging, clinicians can begin the diagnostic process immediately and provide most patients with an answer in less than 30 minutes.
‘For patients, that means faster reassurance when DVT has been ruled out and quicker access to treatment when a clot is identified.’
Chris Chaney, Chief Executive of CW+, said: ‘CW+ funded this work because we believed it could address a real challenge for patients and frontline NHS teams.
‘Through CW Innovation, we help clinicians test promising technologies in real-world healthcare settings. These results demonstrate the potential of AI-enabled diagnostics to improve patient experience and support more efficient models of care across the NHS.’
Following the successful evaluation, the Trust is exploring opportunities to expand the use of the technology, supporting faster diagnosis and improved patient experience.
