Artificial Neural Network Directed Therapy of Severe Obstructive Sleep Apnea

Phase 3

Withdrawn

• Conditions: Sleep Apnea
• Interventions: Other: computer model; Other: Polysomnogram

• Registration Number: NCT01286636
• Lead Sponsor: State University of New York at Buffalo

Brief Summary

The investigators have developed a simple, accurate, and a point-of-care, computer-based clinical decision support system (CDSS) not only to detect the presence of sleep apnea but also to predict its severity. The CDSS is based on deploying an artificial neural network (ANN) derived from anthropomorphic and clinical characteristics.

The investigators hypothesize that patients with severe OSA defined as AHI≥30 can be diagnosed with the use of ANN without undergoing a sleep study, and that empiric management with auto-CPAP has similar outcomes to those who undergo a formal sleep study.

Detailed Description

Not available

Study Timeline

• Study Start: 2011-01
• Study First Submitted: 2011-01-27
• Study First Submitted QC: 2011-01-28
• Study First Posted: 2011-01-31
• Primary Completion: 2015-03
• Study Completion: 2015-06
• Status Verified: 2016-01
• Last Update Submitted: 2016-01-12
• Last Update Posted: 2016-01-13

Recruitment & Eligibility

• Status: WITHDRAWN
• Sex: All
• Target Recruitment: Not specified

Inclusion Criteria

• Must be an adult (≥18 years old)
• Must have symptoms suggestive of OSA, and be considered for sleep study by the sleep specialist provider.

Exclusion Criteria

• Pregnancy or breast feeding
• Patients with severe congestive heart failure (eg, NYHA Class IV, ejection fraction < 35%).
• Patients with end-stage renal disease on hemodialysis
• Patients with CVA, Parkinson, neuromuscular degenerative disease.
• Patient on narcotics.
• Patients with severe lung disease requiring oxygen at night and/or during the day.
• Patient with predominant insomnia or sleep hygiene problems, and who are not considered for PSG by the sleep specialist.

Study & Design

• Study Type: INTERVENTIONAL
• Study Design: PARALLEL

Arm && Interventions

Group	Intervention	Description
artificial neural network	computer model	-
Polysomnogram	Polysomnogram	-

Primary Outcome Measures

Name	Time	Method
To demonstrate that using an ANN directed management of OSA is not inferior to PSG directed management of OSA in terms of sleepiness related functional outcome	6 weeks

Trial Locations

Locations (1)

Veterans Affairs Medical Center in Buffalo; 🇺🇸 Buffalo, New York, United States