Abstract
High risk professions, such as pilots, police officers, and TSA agents, require sustained vigilance over long periods of time and/or under conditions of little sleep. This can lead to performance impairment in occupational tasks. Predicting impaired states before performance decrement manifests is critical to prevent costly and damaging mistakes. We hypothesize that machine learning models developed to analyze indices of eye and face tracking technologies can accurately predict impaired states. To test this we trained 12 types of machine learning algorithms using five methods of feature selection with indices of eye and face tracking to predict the performance of individual subjects during a psychomotor vigilance task completed at 2-h intervals during a 25-h sleep deprivation protocol. Our results show that (1) indices of eye and face tracking are sensitive to physiological and behavioral changes concomitant with impairment; (2) methods of feature selection heavily influence classification performance of machine learning algorithms; and (3) machine learning models using indices of eye and face tracking can correctly predict whether an individual's performance is “normal” or “impaired” with an accuracy up to 81.6%. These methods can be used to develop machine learning based systems intended to prevent operational mishaps due to sleep deprivation by predicting operator impairment, using indices of eye and face tracking.
Highlights
Many professions require workers to perform cognitively challenging tasks for long periods of time and/or under conditions of little sleep
We performed three steps of analyses: first, we confirmed that psychomotor vigilance task (PVT) performance is significantly affected by time awake; second, we investigated changes in the Facial Tracking (FT) and Eye Tracking (ET) indices between “normal” and “impaired” classes to confirm that sleep deprivation-induced performance impairment was reflected in changes in FT and ET indices, and to assess sensitivity of those indices; lastly, we trained 14 different machine learning models with five methods of feature selection to classify subjects performance as “normal” or “impaired” to determine the best machine learning model for prediction of sleep deprivation induced performance impairment
Balanced accuracy was influenced by choice of feature selection method and classifier algorithm; sequential forward selection (SFS) and genetic algorithm (GA) wrapper methods consistently performed better than all-inclusion or filter methods, followed by similar performances from filter methods, and all-inclusion resulting in the worst performance (Figure 4, Table 2)
Summary
Many professions require workers to perform cognitively challenging tasks for long periods of time and/or under conditions of little sleep. Sustained attention to a cognitively demanding task, without sufficient rest leads to fatigue which impairs cognitive performance. This poses a risk to worker safety, public safety, and workplace productivity. Models for Predicting Performance Impairment professions (e.g., military watch positions, air traffic control, sonar operations, etc.), decrements can lead to injury and/or loss of life, be financially costly, and compromise safety and security. Observed performance decrements include increased reaction times to stimuli, greater variability in reaction times, and higher frequency of errors (Basner and Dinges, 2011; Basner et al, 2011). Predicting cognitively impaired states before performance decrements manifest is critical to prevent costly and damaging mistakes
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
