Project investigates safety of AI-powered telemedicine system

An expert from the University’s Centre for Health Technology is part of a research team testing the use of surgery follow-up phone calls made by an ‘AI assistant’.

The half-a-million-pound project, funded by the National Institute for Health Research (NIHR), will investigate whether a Voice Artificial Intelligent Assistant, ‘DORA’, consistently makes the same decisions as an expert clinician, based on conversations with patients who have recently undergone cataract surgery.

The study, Autonomous Telemedicine – Cataract Surgery Follow-up at Two NHS Trusts, is part of work by DORA’s developers, Oxford health tech company Ufonia, to bring the technology into general use in the coming years. The need to alleviate pressure on overstretched clinicians has become all the greater as a result of the COVID-19 pandemic, with almost all elective care being cancelled or postponed in the months following the start of the outbreak, creating a significant backlog. One way to do so is to reduce what the research team call ‘highly stereotyped’ routine work, such as follow-up appointments after cataract surgery, usually conducted face-to-face by a nurse.

Cataract surgery is already the most common operation at most hospitals and in the NHS, with 450,000 procedures carried out each year. The number of patients with cataract is set to double between now and 2050, as the condition is strongly age related.

The care pathway that includes the procedure is ideal for testing new technology like DORA as serious complications are very rare, but surgery requires a post-operative check to confirm that all is well and assess success. Ufonia has developed DORA to be able to place and conduct a phone call to patients a month after surgery, with the aim of replacing the need for an appointment with a clinician.

Dr Edward Meinert, Associate Professor of eHealth in the Centre for Health Technology, is leading the evaluation phase of the project. He said:

“At the moment, this follow-up is done by a nurse, which takes up valuable time and ultimately costs money. Automating this process can also increase flexibility for patients, so it’s a win-win.

“Along with colleagues from the Centre, I’ll be looking at whether it is safe – whether DORA makes the same decisions as clinicians, who will be overseeing every call in the study. But I’ll also be investigating adoption sustainability, whether patients are as happy to talk to DORA as they are to a real person, and whether clinicians feel they can rely on the system.”

During the project DORA will be trialled in addition to the current standard of care for a cohort of patients at Imperial College Healthcare Trust and Oxford University Hospitals NHS Foundation Trust. As well as safety and the attitudes of patients and clinicians, the health economic benefits to patients in reduced time commitment and travel, and the local healthcare system will be assessed. If successful, a proposal will be developed to roll DORA out to all patients at each site, in anticipation of wider NHS deployment in future.

Dr Edward Meinert

Dr Edward Meinert is Associate Professor of eHealth at the University of Plymouth. A chartered engineer, European Engineer, a Fellow of the British Computer Society and Chartered Management Institute, Edward was previously Sir David Cooksey Fellow in Healthcare Translation at the University of Oxford.

With extensive experience in data science, his research focuses on the use of digital technology in healthcare including mobile digital apps, wearables, robotics and clinical artificial intelligence with a primary aim to enable preventive healthcare.

/University Release. The material in this public release comes from the originating organization and may be of a point-in-time nature, edited for clarity, style and length. View in full here.