Facial Depth Maps: Detecting Sleep Apnea with Deep Learning

L. Crownie Majestica , K. Akshay Singh , G. Bhavya , B. Rishitha , Dr. A. Satyanarayana

Obstructive Sleep Apnea, Deep learning, VGG-19

Obstructive Sleep Apnea (OSA) presents a significant health concern, characterized by repetitive airway obstruction during sleep, often leading to symptoms such as snoring, disrupted sleep, and daytime fatigue. Despite its prevalence, OSA diagnosis remains challenging and costly, contributing to under diagnosis and untreated cases. This study explores the potential of deep learning techniques applied to facial depth maps for OSA diagnosis, leveraging the relationship between facial morphology and the condition. By utilizing depth maps, which offer enhanced spatial information compared to 2D images, we aim to improve diagnostic accuracy. Our approach incorporates transfer learning to achieve promising results with limited data, achieving an efficient validation accuracy. We focus on predicting OSA severity, categorizing patients into above-moderate and below-moderate groups based on an apnea-hypopnea index threshold of 15. This research signifies a step toward non-invasive and efficient OSA diagnosis, potentially facilitating timely interventions and improved patient outcomes.

Published Paper   E-Certificate

"Facial Depth Maps: Detecting Sleep Apnea with Deep Learning", IJSDR - International Journal of Scientific Development and Research (www.IJSDR.org), ISSN:2455-2631, Vol.9, Issue 5, page no.1058 - 1063, May-2024, Available :https://ijsdr.org/papers/IJSDR2405141.pdf

Volume 9 Issue 5, May-2024

Pages : 1058 - 1063

Paper Reg. ID: IJSDR_211438

Published Paper Id: IJSDR2405141

Downloads: 000347409

Research Area: Engineering

Country: Hyderabad, Telangana, India

Published Paper PDF: https://ijsdr.org/papers/IJSDR2405141

Published Paper URL: https://ijsdr.org/viewpaperforall?paper=IJSDR2405141

ISSN: 2455-2631 | IMPACT FACTOR: 9.15 Calculated By Google Scholar | ESTD YEAR: 2016

An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.15 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator

Publisher: IJSDR(IJ Publication) Janvi Wave

This work is licensed under a Creative Commons Attribution 4.0 International License and The Open Definition