Diagnosing Obstructive Sleep Apnea Using Electromyography of the Muscles of the Mouth

ConditionsObstructive Sleep Apnea (OSA)OSA Healthy Volunteers Electromyography

Trial statusRecruiting

Trial phaseNot listed

Trial typeObservational

Biological sexAll

Age18-80

SponsorUniversity of California, San Diego

About this trial

The investigators will establish how well a novel, quick, and painless way of measuring muscle activity from the mouth and throat works for detecting sleep apnea. This technology is called transmembraneous electromyography (tmEMG). Leveraging two technologies, a new probe capable of recording muscle activity by lightly touching the muscle, as well as a machine learning model for signal interpretation, the investigators will conduct an initial observational feasibility study in phase 1, followed by a larger observational cohort study in phase 2 to assess the performance of deep learning enhanced tmEMG. The study will address a critical unmet need in sleep apnea diagnostics: the availability of an inexpensive, accurate diagnostic test for screening at point of care.

Eligibility criteria

This trial accepts healthy volunteers

Qualifiers

Age 18 years or older

Subject must have completed a prior sleep test.

Subjects may be recruited if the last prior sleep test either determined that they do not have sleep apnea OR they have moderate to severe sleep apnea. No sleep apnea is defined by AHI < 5; moderate to severe sleep apnea is defined by AHI >= 15.

To be fully eligible, the prior test result needs to be verified via 2-night home sleep test (using the average AHI from the 2 nights). If OSA status (i.e., AHI<5/h or >=15/h) from the 2-night HST vs last prior sleep test is discordant, then participants will be excluded as screen failures.

Disqualifiers

Current use of OSA therapy, including PAP (positive airway pressure) or non-invasive ventilation on a daily basis

Prior uvulopalatopharyngoplasty surgery for sleep apnea.

Inability to install WatchPAT ONE application on smartphone or tablet or unwillingness/inability to use the WatchPAT ONE home sleep study device

Prisoners are excluded due to ethical, legal, and practical concerns

Trial population

Study subjects will be recruited from adult patients who have undergone a sleep study, generally within San Diego County.

Trial design

Design model

Case-control

Time perspective

Prospective

Treatments tested in this trial

Electromyography (EMG) of the Oropharynx
Diagnostic test
The transmembranous electromyography (tmEMG) probe is a disposable single-use bipolar recording device with the sensor configured with two electrodes located at the distal end of the probe in a parallel orientation. Unlike conventional EMG probes that are designed to be inserted into the muscle, the tmEMG probe is placed on the surface of the muscle similar to a surface electrode; the tip of the probe is approximately the size of a ball point pen, which allows us to record from the smaller muscles inside the mouth. Recordings will be taken from the genioglossus and palatoglossus in the mouth bilaterally. Recordings will be taken while the subject performs various maneuvers such as shallow breathing, deep breathing, and pressing their tongue against the side of their mouth.

Treatment groups

60 Participants

are divided into 2 treatment groups

Group A: Healthy Controls1 intervention

Group B: People with Sleep Apnea1 intervention

Trial outcomes

Primary outcomes

AUC (Area Under the Curve)

Diagnostic accuracy, as measured by the area under the receiver operator characteristic curve (AUC), of DL-tmEMG algorithm in detecting moderate to severe OSA.

Time frame

3 months

Secondary outcomes

Delta AUC When Adding Clinical Data

The secondary outcome is to assess the incremental improvement in diagnostic performance, as measured by the change in AUC, when selected clinical variables are incorporated into the tmEMG-based machine learning model. These clinical variables may include demographic characteristics, sleep-related questionnaire scores (e.g., STOP-Bang, Epworth Sleepiness Scale), and relevant medical history. Model performance with and without clinical variables will be compared using appropriate statistical techniques to evaluate the added predictive value.

Time frame

3 months

Sponsors and contacts

Click on the lead sponsor to view all of their trials.

University of California, San Diego

Lead sponsor

National Heart, Lung, and Blood Institute (NHLBI)

Collaborator

Powell Mansfield Inc.