Recipient organisationRoyal Papworth Hospital NHS Foundation Trust
Funding£3.5M
PeriodOct 2021 — Apr 2026
In plain English
AI plain-English summary
A home-monitoring app could warn cystic fibrosis patients ten days before a lung flare-up strikes, giving them time to start treatment early and avoid a hospital visit. This matters because cystic fibrosis patients spend their lives in a cycle of stable periods interrupted by sudden, dangerous lung attacks called acute pulmonary exacerbations. These attacks drive permanent lung damage and premature death. Patients themselves have identified the crushing burden of constant hospital visits as their top research priority. The current model of care is unsustainable. The researchers have already shown that home monitoring can safely cut routine outpatient appointments by half. Now they have built a prototype AI algorithm that predicts impending attacks an average of ten days earlier than conventional care. This project will refine that algorithm using data from a US study, test the app with expert patients, and run a randomised trial across four UK centres. The trial will compare quality of life, lung function, emergency contacts, and antibiotic use between patients who see only their home monitoring data and those who also see the algorithm’s predictions. If successful, the algorithm could achieve CE marking and be rolled out commercially, potentially extending to other chronic lung conditions like COPD and asthma.
View original technical description
Our application builds on our extensive experience of using home-monitoring to manage cystic fibrosis (CF), an exemplar for respiratory conditions where burden of care has been identified by patients as the highest research priority and existing standards of care models recognised as unsustainable. CF is characterised periods of relative stability punctuated by abrupt clinical deterioration known as acute pulmonary exacerbations (APEs) which drive pulmonary inflammation, progressive lung damage, and premature death. In preliminary studies, we have already demonstrated that home monitoring empowers patients and can safely reduce routine outpatient attendance by 50% to the benefit of patients, healthcare providers, and payers. Using home monitoring data from our previous national feasibility study ( SmartCareCF ), we have created a prototype AI-powered predictive algorithm that can forecast the likelihood of an impending APE an average of 10 days earlier than conventional care. We now propose to operationalise this predictive algorithm by: 1) refining the algorithm using data obtained from a US CF Foundation-funded clinical implementation study ( Project Breathe ) in 4 UK CF centres. 2) Optimising the user experience with the application through a series of beta testing evaluations by expert patients led by our PPI lead. 3) Performing a randomised trial (in adults with CF from four leading UK Centres) comparing the impact on patients of seeing home monitoring data alone or in conjunction with the continuous output of the predictive algorithm. We will evaluate the impact of seeing the algorithm output on quality of life (primary output) as well as on physical wellbeing (lung function, weight), emergency contacts to CF Centres, and days of antibiotic treatment. We would anticipate that, by the end of the study, our algorithm will achieve CE marking and be ready for commercial roll out, and could be extended to other chronic respiratory conditions such as COPD and Asthma.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know