An estimated 29.4 million adults in the U.S. suffer from sleep apnea, a major contributor to the “epidemic” of Americans not getting enough sleep, according to the Centers for Disease Control and Prevention.
Sleep apnea—interruptions in breathing that can be caused by obstructions such as the tonsils—is a progressive disorder that can be life-threatening. But before doctors choose a treatment for a patient who doesn’t sleep well, they must figure out whether the patient suffers from sleep apnea or a different disorder. The process of diagnosis, however, tends to be time- and labor-intensive, says Chris Fernandez, co-founder and CEO of EnsoData. The Madison, WI-based startup is developing software that uses machine learning algorithms to help clinicians score sleep data from studies performed in laboratories and at patients’ homes.
“Today, more than 85 percent of sleep clinics rely on a manual scoring process. It can take one hour or more” for a clinician (usually a polysomnographic technologist) to analyze eight or nine hours of sleep data this way, Fernandez says. By contrast, staff at the 50-plus sleep clinics currently using EnsoData’s software can diagnose sleep disorders and other conditions in as little as five minutes, he says.
Investors think EnsoData is on to something. The startup recently announced it raised a $1.5 million funding round, bringing its total seed-stage funding to more than $2 million, Fernandez says.
New York-based Colle Capital led the round. Other participants included two Wisconsin firms, HealthX Ventures and Wisconsin Investment Partners; Chicago-based M25 Group; and Sternhill Associates, which is based in the Boston area. Fernandez says two organizations that use his company’s EnsoSleep software also participated in the recent funding round, though he declined to identify them.
EnsoData will use some of the proceeds from the financing for product development, Fernandez says. The startup plans to double its current headcount, to eight employees, in the next six months, he adds.
About 3.5 million diagnostic sleep tests are ordered in the U.S. each year, according to a report from NovaSom, which helps arrange at-home sleep tests. To prepare for such tests, workers at a sleep clinic fasten sensors to the patient. These devices measure heart and breathing rates, brain waves, oxygen levels, body movements, and other data, the American Society of Sleep Medicine says on its website.
EnsoSleep software uses artificial intelligence (A.I.) and machine learning techniques to assess patients’ sleep tests—specifically, whether the information collected by sensors suggests a patient likely has sleep apnea or another disorder, Fernandez says. As the software processes more data over time, its ability to accurately diagnose disorders improves, he says.
“Our A.I. approach … and the data network effect are improving our software with every single customer we add,” Fernandez says.
Fernandez co-founded EnsoData in 2015, along with company president Sam Rusk and chief technical officer Nick Glattard. About a year ago, the FDA cleared EnsoSleep for sale and clinical use in the U.S.
EnsoData has identified “immediate opportunities” to market its software outside the U.S., Fernandez says. He declined to say which countries the company is considering expanding into, but says they all “have some level of medical device and data privacy regulation.” EnsoData will need to ensure it’s in compliance with those regulations before it begins selling
