Screening and management of obstructive sleep apnea (OSA) and daytime sleepiness in professional drivers in Tunisia: A protocol study using machine learning

Author:

MSAAD Sameh1,GARGOURI Rahma1,KAMMOUN Nesrine2,KHEMEKHEM Rim1,ABID Narjes2,FEHRI Sonia2,KALLEL Kaouther2,KAMMOUN Rim2,GHARBI Leila DOUIK EL2,Maalej Sonia2,Bouhamed Heni1,Abdelghani Ahmed3,Aichaouia Chiraz2,Turki Mohamed1,KAMMOUN Samy1

Affiliation:

1. University of Sfax, Sfax

2. University of Tunis EL Manar, Tunis

3. Univrsity of Sousse, Souse

Abstract

Abstract

Background Professional drivers have high rates of Obstructive Sleep Apnea (OSA). This study aims to determine the prevalence of OSA and Excessive Daytime Sleepiness (EDS) and identify their risk factors among a large representative sample of professional drivers in Tunisia. We will also evaluate the risk of accidents associated with OSA and EDS before and after treatment. Methods This will be a population-based and prospective study about 3000 professional drivers. Participants will receive a structured questionnaire to evaluate five main outcomes: the likelihood of OSA, EDS, drowsy driving, related sleepiness near misses and accidents, as well as work productivity. Validated self-report measures will be used to evaluate these outcomes. Participants suspected of having OSA and/or EDS will undergo sleep laboratory investigations, including sleep study. Participants who have moderate-to-severe OSA combined with EDS will be recommended CPAP treatment. After one year of follow-up, all participants will be re-evaluated with self-report questionnaires. For those treated with CPAP, they will undergo the Maintenance of Wakefulness Test (MWT) and adherence, tolerance, and consistency of CPAP usage will be recorded. We will also use machine learning models. Among the models we will evaluate: Random Forests (RF), XGBoost, and Deep Neural Networks (DNN). The data will be split into two sets: a training set and a test set. The training set will be used to train the machine learning models. The test set will be used to evaluate the models' performances. We will compare the performances of the different machine learning models using standard evaluation metrics such as accuracy, recall, and F1-score. Conclusion Our results will pave the way for the creation of a clinical screening instrument that can identify sleep-wake disturbances in professional drivers. This is likely to have a significant impact on the legal regulations concerning driving fitness and road safety.

Publisher

Springer Science and Business Media LLC

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3