Loughborough University has developed an AI model, aimed at tackling “healthcare challenges faced by people with learning disabilities and multiple health conditions”, with the tool being applied at Leicestershire Partnership NHS Trust.
According to a team of computer scientists that have been working on the DECODE project, the model has the ability to “predict how long a person with a learning disability is likely to stay in hospital,” aiming to improve care and resource planning.
The AI model was trained on GP and hospital data from over 9,600 patients with learning disabilities and multiple health conditions in Wales and was then used to analyse key data to help identify “reasons for hospitalisations and health patterns among people with learning disabilities and multiple health conditions”.
The researchers highlight that the model is “76% effective in distinguishing between patients likely to have prolonged hospital stays and those who would be discharged sooner”.
The learning is now being applied at Leicestershire Partnership NHS, with consultant psychiatrist with Leicestershire Partnership and the DECODE co-principal investigator, Dr Satheesh Gangadharan, noting: “We are in the process of applying this knowledge into practice as well as sharing it widely. While hospital care is an important part of healthcare provision, we are exploring ways to minimise the need for hospitalisation by exploring where health interventions could be delivered earlier and people with learning disabilities could be engaged in their care better.”
The researchers at Loughborough University are planning to apply the same AI model to English datasets to “assess whether similar patterns emerge across different populations” while also using the information gained from their study to “support the NHS in developing risk prediction algorithms” for decision-making.
AI in healthcare: the wider trend
An AI transcript tool known as CLEARNotes is being piloted at The Royal Wolverhampton NHS Trust. It reportedly generates a structured discussion summary from conversations between doctors and patients, in hopes of reducing the administrative burden for clinicians.
For a recent HTN interview, we caught up with Michael Wornow, a computer science PhD student at Stanford University, to discuss some of his most recent projects on developing and operationalising AI models in healthcare. He mainly concentrated on his involvement with the research on Advancing Responsible Healthcare AI with Longitudinal EHR Datasets, including the introduction of new datasets and key findings from the study.
The UK government recently published a response to a report from the Regulatory Horizons Council (RHC) about the regulation of AI as a medical device (AIaMD). All 15 of its recommendations were accepted, while the government also gave insights into current work and next steps in this space. Key areas of focus in the recommendations made in the report cover regulatory capacity and capability.
Learn more about AI and data in health and care with all the latest news from HTN.
