Join Farnaz Hajipour from the University of Manitoba for her talk on Thursday, February 13th about “Comparing Regularized Logistic Regression and Random Forest Approaches for Daytime Diagnosis of Obstructive Sleep Apnea”.


Obstructive sleep apnea (OSA) is a prevalent yet under diagnosed health problem. Assessment of OSA is currently based on sleep studies that are time-consuming and expensive. Developing technologies for quick OSA screening is momentous. Studies have been shown that the upper airways (UA) structural and physiological changes can alter the tracheal breathing sounds (TBS) characteristics. We hypothesize the TBS analysis during wakefulness correlate with the severity of OSA and could represent physiological characteristics of UA. In the context of TBS analysis, it is possible to extract a considerable number of features from data. A major challenge in high-dimensional data analysis is related to building parsimonious models and removing variables that do not add any information to our model. This study aims to assess the utility of two important machine-learning techniques to classify subjects with OSA using their daytime TBS. We evaluate and compare the performance of the Random Forest (RF) and Regularized Logistic Regression (LR) as feature selection tools and classification approaches for wakefulness OSA screening. The findings of this study may help in enhancing the current OSA algorithms by providing a non-time-consuming and less expensive method to stratify the severity of OSA patients in a fast but more precise way.

DATE: Thursday, February 13th, 2020

WHERE: 111 Armes Building

WHEN: 3:45pm


Learn more here
Feb 4, 2020