AI used to predict COPD flare ups from urine samples
· News-MedicalResearchers have used artificial intelligence (AI) to analyse patient urine samples and predict when symptoms of chronic obstructive pulmonary disease (COPD) will flare up according to a study published today (Wednesday) in ERJ Open Research.
The patients taking part in the study carried out a simple daily dipstick test on their urine and sent their results to researchers using their mobile phones.
Using AI to analyse the results, researchers were able to 'forecast' a deterioration in symptoms one week in advance. This could make it possible to take steps, such as altering treatment, to minimise or even prevent a flare up.
COPD, which includes emphysema and chronic bronchitis, is a serious and long-term lung condition. According to the World Health Organization, COPD is the third leading cause of death worldwide. A flare up in symptoms, such as difficulty breathing and coughing, is known as an exacerbation.
The study was led by Professor Chris Brightling from the University of Leicester, UK, which is part of the National Institute for Health and Social Care Leicester Biomedical Research Centre. He said: "COPD exacerbations are when someone with COPD becomes very ill and needs additional treatment either at home or in hospital. The current treatments are reactive to a severe illness. It would be better if we could predict an attack before it happens and then personalise treatment to either prevent the attack or reduce its impact. We wanted to develop a predictive test that would act like a personal weather forecast of an impending flare-up."
The researchers began by analysing urine samples from a group of 55 people with COPD and looking for any changes in the make-up of their urine that preceded a deterioration in symptoms. This helped them to identify a set of 'biomarkers' – molecules that tend to change when COPD is worsening.
The researchers used a type of AI called an artificial neural network to look for changes in the levels of these biomarkers and predict when a patient was going to experience a flare up in COPD symptoms.
They found this AI analysis could accurately predict a flare up around seven days before any symptoms appeared.
The advantage of sampling urine is that it's relatively quick and easy for patients to do at home on a daily basis. We need to do more work to refine the AI algorithm with data from a bigger group of patients. We hope this will allow us to create AI testing for COPD patients that will learn what is 'normal' for each person and forecast a flare up in symptoms. Patients' care could then be adapted, for example they might need further testing or treatment, or they might be able to limit their exposure to triggers like pollution or pollen." Professor Chris Brightling, University of Leicester, UKOur study first explored many substances in urine samples from people with COPD during a flare up and when they were stable. We found that a small number of these substances could identify a flare up. We then followed up a group of people with COPD and tested five substances daily. This allowed us to develop the risk prediction or forecasting AI-tool. We found the AI tool could reliably predict a flare up in symptoms seven days prior to a diagnosis.
Professor Apostolos Bossios from the Karolinska Institutet and Karolinska University Hospital, Stockholm, Sweden, is Head of the European Respiratory Society's airway disease assembly and was not involved in the research. He said: "COPD is a common and serious condition. There is no cure for COPD, so monitoring and treatment is crucial for helping patients stay well enough to carry out their normal day-to-day activities.
"When COPD symptoms flare up, it can lead to permanent deterioration, so we want to do all we can to prevent or minimise flare ups. This research is promising because it suggests we can use AI analysis of urine samples to predict a flare up before it starts. If it proves successful in the longer term, this testing could make sure patients get the treatment and care they need to reduce symptom flare-ups as quickly as possible."
Source:
Journal reference: