Novel Techniques to Predict Obstructive Seep Apnea Severity during Daytime (Wakefulness) using Machine-Learning Algorithms
Zahra Moussavi
October 5, 2022, Wednesday, 3:00 PM - 4:00 PM EST
Obstructive sleep apnea (OSA) is an underdiagnosed common disorder. Undiagnosed OSA, in particular, increases the perioperative morbidity and mortality risks for OSA patients undergoing surgery requiring full anesthesia. OSA screening using the gold standard, Polysomnography (PSG), is expensive, in need of expert skills and time-consuming.

In this talk, I will present the research on developing objective and accurate alternative tools for screening OSA during daytime (wakefulness) by a few minutes of breathing sounds analysis. In particular, I will explain our new algorithm, AWakeOSA, that extracts an optimized set (3-4) of breathing sound features specific to each anthropometric feature (i.e. age, sex, etc.), and uses a two-level diagnostic classification using machine learning to determine OSA severity for each individual.
Dr. Moussavi has published a total of 286 scientific publications including: 5 patents, 3 chapter-books, 92 peer reviewed full journal papers in high impact journals, 10 brief technical journal papers, 116 peer reviewed conference full papers, and 60 abstracts with an overall >4242 citations. She is recognized for her work on two major areas: 1) respiratory acoustics signal processing and its application on sleep apnea detection; the series of publications and patents in that area has resulted in several national and international research collaborations; 2) the design of innovative immersive virtual reality navigational environments for Alzheimer's diagnosis and its treatment by non-pharmacological means; this line of research has resulted in two international collaboration and funding. Excluding regular conference talks, she has given 88 invited scientific talks (42 outside of Canada) including 15 keynote speaker seminars at national and international conferences. In addition, she has given 23 outreach scientific talks for public including 2 TEDx talks. On my spare time, I also write science articles for public; so far four articles: 2 in Conversation and 2 in Phebe Journal. Overall, she has trained 11 pdf and research associates, 20 Ph.D, 34 M.Sc., and >100 undergraduate students, and 12 part-time research assistants. Her current team includes 3 pdf, 1 Research Associate, 4 Ph.D., 7 M.Sc and 8 undergraduate students and 4 part-time research assistants.