Our Quest for Avoiding Exacerbations with the Help of AI
The ability to identify the difference between having a “bad day” versus oncoming exacerbation could be life-saving for many respiratory patients. There are many different definitions of exacerbation but it is often referred to as the sustained worsening of symptoms beyond normal day-to-day variations that requires acute intervention by a doctor. Unfortunately, in some cases, the cause for an exacerbation remains unknown and the patient fails to recognize any signs that could have warned them about the threatening situation.
The good news is, that a set of recent university studies indicate that upcoming exacerbations could actually be detected prior to their occurrence through the help of electronic home monitoring. This evidence has been generated for several respiratory conditions, let us highlight three publications that we find highly interesting:
The 2018 publication titled “Spirometric changes during exacerbations of COPD: a post hoc analysis of the WISDOM trial.” highlights the importance of daily lung function testing to help predict exacerbations for COPD patients. The German researchers involved in the trial concluded that mean lung function (FEV1) begins to decline about 2 weeks prior to the first reported symptoms of an exacerbation and does not recover to pre-exacerbation levels 8 weeks after the event.
A study that was done in 2016 in the UK on IPF, titled “Daily Home Spirometry: An Effective Tool for Detecting Progression in Idiopathic Pulmonary Fibrosis”, calls attention to the importance of daily spirometry measurement. “The rate of change in patient-administered daily FVC over 3 months was strongly predictive of subsequent disease progression at 1 year as defined by a composite of >10% change in hospital-measured FVC or death” or, put more placatively, “data suggested that home spirometry could be used as a means for identifying acute exacerbations of IPF at an early stage”.
Regular monitoring of lung function at home also proved to be effective with children diagnosed with Cystic Fibrosis. A 2017 study by Dutch researchers, consisting of 49 children who monitored their condition at home three times a week showed significant results. The study concluded that “The combination of home monitor FEV1 and a Respiratory Symptoms Score can be helpful to predict an exacerbation in children with CF at an early stage”. It was shown that respiratory symptoms increased 2 weeks prior to the exacerbation.
All studies highlight that exacerbations are preceded by measurable deviations in lung functions and symptoms. It has also been shown that home spirometry can be an effective tool to measure these deviations. All these observations have been done retrospectively, after the exacerbation occurred. At NuvoAir, we have the vision that technology can soon help to actually predict risk situations before they occur. Therefore, we are already today working on algorithms to help patients and doctors to detect worsening of conditions at an early stage.
As Artificial Intelligence continues to play an actively increasing role in our everyday lives, we believe it will soon be a powerful tool to assist individuals with respiratory diseases. Patients need to understand when they are approaching a risky situation and take appropriate measures to manage their health. Throughout the past couple of years, we have started analyzing patterns from our database of over 500,000 spirometry tests and are combining these insights with additional data points from sensors and questionnaires. Ultimately with the goal of creating an AI assistant that understands a combination of different endpoints and that can help patients and doctors to identify respiratory health risks at an early stage.
Although predicting exacerbations and deterioration remain difficult and require a lot of expertise, solving this equation to help millions of patients around the world is what keeps us motivated. We are collaborating with some leading researchers in this field to make this vision come true. Don’t hesitate to contact us if you want to learn more or want to get involved.