Affordable Access

Publisher Website

A Benchmark on Automatic Obstructive Sleep Apnea Screening Algorithms in Children

Authors
Publisher
Elsevier B.V.
Volume
35
Identifiers
DOI: 10.1016/j.procs.2014.08.156
Keywords
  • Apnea Screening
  • Supervised Learning
  • Classification
  • Benchmark
Disciplines
  • Computer Science
  • Medicine

Abstract

Abstract Sleep Disordered Breathing (SDB) are a group of diseases that affect the normal respiratory function during sleep, from primary snoring to obstructive sleep apnea (OSA). Children affected by OSA may develop growing disorders and even long-term cognitive disadvantages. However, once they have been diagnosed, treatment is effective in most of the cases improving their quality of life and avoiding consequences in their cognitive development. Although, several models have been reported to be good automatic OSA predictor in adults; no study have been conducted to test whether these models holds when predicting children’ OSA or not. Our study uses the largest data base of polysomnogram data in Children under 15 years old. We benchmarked the three best methodologies reported on the literature. Our results show that these models’ predictive power is drastically reduced when applied to Children. We present the bases to develop new algorithms which can perform automatic OSA screening in Children.

There are no comments yet on this publication. Be the first to share your thoughts.