Project Details
Description
The accurate diagnosis of OSA is based on the results from an in-laboratory sleep study, which is often only available in sophisticated medical centers. It is not surprising that 75-80% of the patients with OSA remain undiagnosed and develop life-threatening comorbidities. The socio-economic impact is difficult to determine because of the high rate of comorbidities of OSA. According to a study in 2008, the National Safety Council attributes 810,000 motor vehicle crashes over a period of 8 years with 1,400 fatalities and a cost of 15.9 billion dollars to sleep apnea. The same study points out that the mean medical costs for a patient with OSA doubles when compared with a patient from a matched control group without OSA.
Since the majority of people with OSA remain undiagnosed and untreated, it would be important to have a cost-efficient device to diagnose and treat OSA. The key physiological parameters that should be evaluated for a diagnosis are blood oxygen saturation, (oro-nasal) airflow, breathing rate, and breathing effort. Although devices developed for home sleep testing often acquire many more than the key physiological parameters, they rely on a clinician to interpret the results. Furthermore, the devices cannot identify the location of the airway obstruction or have the option for treatment incorporated.
In contrast to existing technology, our newly developed wearable device has implemented all sensors within a 12 cm wide necklace to measure the four necessary parameters to diagnose OSA. Those parameters include: blood oxygen saturation, airflow, breathing rate, and breathing effort. Blood oxygen saturation, heart rate are measured optically with a 3.4 x 5.7 mm2 chip (e.g. MAX30105). Airflow, breathing rate, obstruction during inhalation or exhalation, and the site of obstruction are determined from the acoustic fingerprint from an array of piezoelectric microphones. Both sets of sensors are commercially available and affordable. Data, at-present, is transmitted via a USB cable from a credit-card-sized single chip computer to the master computer. Data are then processed and used in machine learning models to identify sleep events such hypopnea and apnea and predict their occurrence even before they occur. The expected cost of our device at economies of scale is estimated to be below $100. The goal for a one-year funding period is to perform user studies that will inform the design, fabrication and validation of a scalable and manufacturable device.
Status | Finished |
---|---|
Effective start/end date | 8/1/18 → 7/31/19 |
Funding
- Cleveland Clinic Lerner College of Medicine of CWRU (1098SUB // U54HL119810)
- National Heart, Lung, and Blood Institute (1098SUB // U54HL119810)
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.