UK hospital deploys predictive AI for direct care of COPD patients (Guest blog by Lenus Health)
Lenus Health CEO, Paul McGinness on the launch of a world-first study at NHS Greater Glasgow and Clyde aiming to transform management of lung disease.
Chronic obstructive pulmonary disease (COPD) is a progressive and preventable disease that affects around 1.2 million people in the UK and is the second most common cause of emergency hospital admissions. The annual economic burden of COPD on the NHS is estimated as £1.9 billion with treatment following exacerbations of symptoms accounting for 70% of these costs.
A clinical investigation launched recently will explore how insights from artificial intelligence (AI) can improve care and prevent emergency hospital admissions for people with the chronic lung disease.
Supported by a £1.2 million NHS Artificial Intelligence in Health and Care Award, the ‘DYNAMIC-AI’ clinical investigation is a 12-month feasibility study underway in COPD patients at NHS Greater Glasgow and Clyde (NHSGGC) in Scotland.
The study will use machine learning in live point-of-care workflows to identify patients at highest risk of adverse events and support the proactive delivery of guideline-based care.
Direct care of patients
The project between Lenus Health and NHSGGC is thought to be the first to operationalise predictive AI insight for COPD into live clinical practice.
With the ageing population and rising prevalence and complexity of long-term conditions, clinicians are overwhelmed with data that they don’t have the capacity to review. Assistive technologies can provide prioritised insights from patient data, allowing clinicians to optimise preventative management and freeing up time to focus on human interactions.
Moving from rule-based care to machine learning models
Data scientists and engineers at Lenus Health have developed four machine learning models to identify COPD patients at risk of adverse events. The AI algorithms have been trained using close to one million data points from historical electronic health records from a de-identified cohort of more than 55,000 patients with COPD resident in NHSGGC. They use more than 80 data points to support the delivery of risk scores, significantly more than in a traditional rule-based system, which is known to cause frequent false alarms leading to clinician alarm fatigue.
Under the Dynamic-AI study, clinical care teams will be provided with actionable insights from the models to use in multi-disciplinary team (MDT) reviews. By identifying high risk patients, they can be offered proactive, preventative care to avoid the COPD symptom flare-ups that currently cause one in eight emergency hospital admissions.
Up until now, AI models have been used retrospectively in cohorts to provide predictions looking back. We believe this is the first time AI-derived predictive scores will be used directly within the day-to-day clinical workflow in COPD care. This can help us transform to a preventative, predictive and proactive care model that improves outcomes for patients and relieves pressures on the care system.
Addressing equality issues
COPD disproportionately affects deprived populations, with the prevalence in the most deprived 10% of areas of the UK estimated to be almost double that of the least deprived 10%.
The project’s work on fairness aims to identify biases held in health data and ensure models perform appropriately across age groups, gender, deprivation categories and ethnicity. This has the potential to improve access to healthcare to people living in socio-economically disadvantaged areas.
We anticipate that the approach, methodology and wider learnings from this exemplar study will accelerate the application of AI across additional long-term conditions for a strategic approach to addressing health inequalities, unprecedented acute care demands, and expanding economic costs.
Get our tech and innovation insights straight to your inbox
Sign-up to get the latest updates and opportunities from our Technology and Innovation and AI programmes.
Authors
