Can WatchPAT’s Raw Data Predict Cardiovascular Outcomes? New Research Partnership Aims to Find Out

Itamar Medical Ltd has entered into a research collaboration agreement with Clalit Research Institute (CRI), a subsidiary of Clalit Health Services (CHS), Israel’s largest state-mandated health service organization.

Under the agreement, Itamar and CRI will collaborate to conduct a study designed to determine if data from the WatchPAT signal can help predict cardiovascular health outcomes. The study will use artificial intelligence (AI) tools to compare predictive models for different outcomes, such as cardiovascular disease, atrial fibrillation, and congestive heart failure. Approximately 50,000 patients within the CHS network have undergone WatchPAT testing over the past 10 years. As a state-mandated health services organization, CRI has access to comprehensive patient health medical records and demographic and longitudinal follow-up data going back 20 years for most of these patients with WatchPAT data.

“The association between sleep-disordered breathing and serious health outcomes is well documented, yet data typically collected from overnight sleep studies are summarized into a small number of reported metrics,” says Gilad Glick, president and CEO of Itamar Medical, in a release. “We believe that the data contained in the WatchPAT raw signal recorded over the entire night has tremendous potential not only to aid in the diagnosis of sleep apnea and help phenotype patients, but also to predict serious cardiovascular health outcomes. As the largest state-mandated health service organization in Israel, CRI has access to a trove of patient data that will, for the first time, allow us to explore the predictive potential of our WatchPAT technology. We look forward to collaborating with CRI to explore how WatchPAT may play a role in improving healthcare outcomes across a wide range of indications.”

[RELATED: Artificial Intelligence in Sleep Medicine Could Reveal New Population Health Insights]

The study will be conducted in four phases that include data acquisition; extracting appropriate study populations for each healthcare outcome evaluated; baseline modeling using electronic medical record (EMR) data and published literature; and WatchPAT signal modeling using AI and machine learning to integrate WatchPAT signal data and EMR data. The study is anticipated to take two years.

“Signal processing can capture ‘hidden’ objective features connected to a variety of diseases. We are excited to evaluate its potential to predict a wide array of cardiovascular diseases. If the model will prove itself; it will be a huge potential to expand this model globally to other healthcare systems especially in poor countries with no well-established EMR systems,” says Ran Balicer, MD, MPH, the director of Clalit Research Institute, in a release.

from Sleep Review https://ift.tt/2BH4Rmv