
Researchers in Europe have used artificial intelligence (AI) to analyze urine samples to predict when symptoms of COPD will flare up.
According a study published in ERJ Open Research, patients carried out a daily dipstick test on their urine and sent results to researchers via their mobile phones. The researchers then used AI to analyze the results and forecast a deterioration in symptoms one week in advance.
“The current (COPD) treatments are reactive to a severe illness. It would be better if we could predict an attack before it happens and then personalize treatment to either prevent the attack or reduce its impact,” said lead researcher Chris Brightling, PhD, in a news release.
The researchers analyzed urine samples from 55 people with COPD and looked for any changes in the samples that preceded a deterioration in COPD symptoms. They used the information to identify a set of biomarkers in the form of molecules that tend to change when COPD gets worse.
From there, they developed a urine test with help from Global Access Diagnostics, Bedford, U.K., that measures five of the biomarkers. They took the tests to 105 COPD patients from two different U.K. hospitals and asked them to test their urine once a day for six months, again sending the results by mobile phone.
Using an artificial neural network, the researchers analyzed results to look for changes in the levels of the biomarkers and predict when flare-ups in COPD symptoms might occur.
“This allowed us to develop the risk prediction or forecasting AI tool,” said Dr. Brightling, who is professor at the University of Leicester, U.K. “We found the AI tool could reliably predict a flare-up in symptoms seven days prior to diagnosis. The advantage of sampling urine is that it’s relatively quick and easy for patients to do at home on a daily basis.”
Dr. Brightling said the next step is to continue working on refining the 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,” he said. “Patients’ care could 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.”