Loading...

Evaluating the Effect of CPAP Pressure in Patient with Obstructive Sleep Apnea Using Heart Rate Variability

Ahmadi Mousavi, Mohammad | 2020

941 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 53141 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Vosoughi, Gholamreza; Arjmand, Navid
  7. Abstract:
  8. Continuous positive airway pressure (CPAP) is a standard treatment for patients with obstructive sleep apnea (OSA), a sleep-related breathing disorder characterized by full or partial occlusion of the upper airway during sleep. CPAP pressure adjust by a sleep technologist during attended laboratory polysomnography (PSG) to eliminate obstructive respiratory-related events (apneas, hypopneas). Because of changes in environment and lifestyle patients needed new adjustments for the device that brings discomfort and extra money for treatment. In recent, lots of methods have been proposed to replace PSG and minimized the number of biological signals to detect apnea, best results came from deep learning framework. But last studies only tested on normal breathing in the absence of breathing device and they didn’t detect all kinds of apnea. In this study, we propose a method for the automated evaluation of CPAP pressure from a single-lead electrocardiogram (ECG) using a convolutional neural network (CNN) that optimized with a genetic algorithm. For training and evaluation of the CNN model, an experimental test designed to collect data from 13 patients while specialists setting the pressure. This framework detected all kinds of apnea with 66.18% accuracy that led to evaluate CPAP pressure with 88.75% accuracy. This evaluation with preventing extra referral for new pressure settings can reduce treatment costs and increase patient comfort
  9. Keywords:
  10. Deep Learning ; Genetic Algorithm ; Apnea ; Electrocardiogram ; Sleep Disorder Breathing (SDB) ; Continuous Positive Airway Pressure (CPAP) ; Electrocardiography

 Digital Object List

 Bookmark

...see more