C Dunbar, P Nguyen, A Guyett, K Nguyen, K Bickley, A Reynolds, M Hughes, H Scott, R Adams, L Lack, P Catcheside, J Cori, M Howard, C Anderson, N Lovato, A Vakulin
{"title":"P071 Simple Vestibular-Occular Motor Assessment as a Predictor of Driving Performance Vulnerability following extended Wakefulness","authors":"C Dunbar, P Nguyen, A Guyett, K Nguyen, K Bickley, A Reynolds, M Hughes, H Scott, R Adams, L Lack, P Catcheside, J Cori, M Howard, C Anderson, N Lovato, A Vakulin","doi":"10.1093/sleepadvances/zpad035.156","DOIUrl":null,"url":null,"abstract":"Abstract Introduction Driver fatigue is a significant contributor to road crashes, but identifying individuals at driving risk is challenging. We examined the potential of simple baseline vestibular ocular motor system (VOMS) assessment via virtual reality goggles to predict subsequent vulnerability to driving simulator impairment following extended wakefulness. Methods 49 individuals (Mean±SD Age 32.6±12.9, 45% Males) underwent 9hr baseline sleep opportunity followed by approximately ~29hrs extended wakefulness with five 60min driving assessments. Cluster analysis classified drivers into vulnerable (n=17) or resistant (n=32) groups based on their worst steering deviation and number of crashes from driving tests. Baseline VOMS were performed ~10mins prior to the first three drives (1, 7 and 13hrs of wakefulness). XGBoost machine learning model was trained using baseline VOMs features to predict vulnerable vs resistant groups from driving tests 4 and 5 (19 and 25hrs of wakefulness) Model performance was evaluated using 5-fold cross-validation approach using ROC analysis. Results XGBoost machine learning ranked all 70 VOMS metrics on their importance in predicting vulnerable vs resistant groups. Top 10 VOMs metrics assessed during baseline non-sleep deprived tests demonstrated a strong ability to predict the driver's performance following extended wakefulness, differentiating between the vulnerable vs resistant groups (AUC 0.73 [95%CI 0.61-0.83, p<0.001]). Conclusion VOMs tests conducted at baseline holds promise for predicting future driving impairment. This approach has the potential to be highly valuable in determining an individual's fitness to drive. Future validation in independent samples, sleep disordered population and in-field on-road testing are needed to confirm these promising findings.","PeriodicalId":21861,"journal":{"name":"SLEEP Advances","volume":"28 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"SLEEP Advances","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1093/sleepadvances/zpad035.156","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Abstract Introduction Driver fatigue is a significant contributor to road crashes, but identifying individuals at driving risk is challenging. We examined the potential of simple baseline vestibular ocular motor system (VOMS) assessment via virtual reality goggles to predict subsequent vulnerability to driving simulator impairment following extended wakefulness. Methods 49 individuals (Mean±SD Age 32.6±12.9, 45% Males) underwent 9hr baseline sleep opportunity followed by approximately ~29hrs extended wakefulness with five 60min driving assessments. Cluster analysis classified drivers into vulnerable (n=17) or resistant (n=32) groups based on their worst steering deviation and number of crashes from driving tests. Baseline VOMS were performed ~10mins prior to the first three drives (1, 7 and 13hrs of wakefulness). XGBoost machine learning model was trained using baseline VOMs features to predict vulnerable vs resistant groups from driving tests 4 and 5 (19 and 25hrs of wakefulness) Model performance was evaluated using 5-fold cross-validation approach using ROC analysis. Results XGBoost machine learning ranked all 70 VOMS metrics on their importance in predicting vulnerable vs resistant groups. Top 10 VOMs metrics assessed during baseline non-sleep deprived tests demonstrated a strong ability to predict the driver's performance following extended wakefulness, differentiating between the vulnerable vs resistant groups (AUC 0.73 [95%CI 0.61-0.83, p<0.001]). Conclusion VOMs tests conducted at baseline holds promise for predicting future driving impairment. This approach has the potential to be highly valuable in determining an individual's fitness to drive. Future validation in independent samples, sleep disordered population and in-field on-road testing are needed to confirm these promising findings.