References of "François, Clémentine"
     in
Bookmark and Share    
Peer Reviewed
See detailValidation of a new automatic drowsiness quantification system for drivers
Wertz, Jérôme ULg; François, Clémentine ULg; Verly, Jacques ULg

Poster (2014, July)

Drowsiness is a major cause of various types of accidents, and particularly of driving accidents. Scientific studies report that drowsiness would be responsible for 20 to 30% of driving accidents ... [more ▼]

Drowsiness is a major cause of various types of accidents, and particularly of driving accidents. Scientific studies report that drowsiness would be responsible for 20 to 30% of driving accidents. Drowsiness can result from, among others, sleep deprivation, sleep disorders, alcohol, some medications, or performing a monotonous task. While all persons are likely to be drowsy at some point during the day, some persons are more prone than others to being drowsy at almost any time of the day; e.g. 6 to 11% of the population suffers from severe chronic excessive daytime sleepiness. Three main classes of methods can be used to characterize the level of drowsiness of a driver without disturbing him. These classes are respectively related to vehicle behavior (e.g. via lateral movements), driver behavior (e.g. via steering wheel movements), and driver physiological state (e.g. via eye movements). Since drowsiness is a physiological state, it seems particularly meaningful to use physiology-based methods to characterize it. Among these, the most significant ones rely on polysomnography and/or oculography. Polysomnography is viewed by some practitioners as the reference in the domain, but it is very sensitive to artifacts, and it is not very practical for use while driving. Ocular parameters are recognized to be good and reliable physiological indicators of drowsiness, and, thus, oculography seems to be the most sensible way to characterize drowsiness in practice. We have thus developed an experimental, fully automatic drowsiness monitoring system (software/algorithms) based on the physiological state of a person. This system uses ocular parameters extracted from images of the eye (i.e. photooculography) to determine a level of drowsiness on a continuous numerical scale from 0 to 10, with 0 corresponding to "very awake" (or "very vigilant") and 10 to "very drowsy". The ultimate goal of this system is to prevent drowsiness-related accidents for driving and other applications. The reported study shows that our system exhibits promising capability for road safety. Fourteen healthy volunteers (7 M, 7 F, mean age 23.7, range 21-33 years) participated in an experiment in a driving simulator, where they were asked to perform three driving sessions/runs (two of 45 minutes and one of 60 minutes) in different sleep-deprivation conditions (with up to 28 hours of complete sleep deprivation). During each session, we recorded both a high frame rate video of one eye and a set of driving parameters. Subsequently, for each successive minute in the session, we used our algorithms to extract ocular parameters from the video images and to produce a level of drowsiness, and we computed the standard deviation of lateral position (SDLP) from the driving parameters. The results show (1) that the (computed) SDLP increases when the (computed) level of drowsiness increases, and (2) that the level of drowsiness increases when the level of sleep deprivation increases. These results indicate that our algorithms for producing a level of drowsiness work in a meaningful way. The experiment protocol was approved by the Ethics Committee of our university. [less ▲]

Detailed reference viewed: 44 (13 ULg)
Peer Reviewed
See detailDetection of eye movements specific to drowsiness and their relation with subjective assessment: A cognitive ergonomic approach.
Blavier, Adelaïde ULg; Montagnino, Cédric; Wertz, Jérôme ULg et al

Poster (2014, May 27)

Drowsiness is one of the major factors explaining accidents, particularly in traffic accidents but also in work situations with serious consequences (e.g. medicine). The drowsiness may be assessed by ... [more ▼]

Drowsiness is one of the major factors explaining accidents, particularly in traffic accidents but also in work situations with serious consequences (e.g. medicine). The drowsiness may be assessed by diverse measures that vary from physiological and unconscious data (e.g. EEG) to subjective and conscious evaluation. In their daily life, people are used to evaluate their drowsiness by subjective assessment and research observes a great inter-individual variation in this evaluation. Moreover, the subjective evaluation is dependent on the situation and the risk perceived by the person (e.g., new versus usual situations, simple versus complex environments, etc.). In this theoretical context, our purpose was to investigate the links between 1) objective performance (reaction time) measured by a psychomotor vigilance task (PVT), 2) data from eye movements and 3) subjective assessment of drowsiness (measured with Karolinska Sleepiness Scale, KSS). 12 persons, aged from 20 to 56, participated individually. They were asked to respect a 60% sleep deprivation during the night before the experiment and to not drink any energy drinks the day of the experiment. The experiment was conducted between 1 and 3 PM after a heavy lunch in order to increase the circadian effect. Moreover, the temperature of the room was 25° in order to increase drowsiness. Each participant was asked to perform 4 PVT. However, although 100% of participants performed the first two PVT, only 66% were able to perform the third PVT and only 33% of participants performed the fourth and last PVT. Our results showed an effect of time on objective performance, eye movements and subjective assessment of drowsiness in PVT 1 and 2: significant increase of reaction time, increase of eye closure and perclos, reduction of pupil diameter and increase of subjective drowsiness estimation (KSS). Only the frequency and duration of blinks remained constant across time. In PVT 3 and 4, data from eye movements and objective performance (reaction time) did not vary anymore across the time. Only the subjective estimation of drowsiness (KSS) continued to increase. Furthermore, subjective estimation of drowsiness was differently correlated with eye movements and objective performance across the PVT; it was significantly correlated with reaction time (PVT 1,2,3), blink frequency and duration (PVT 1,2,3), perclos (PVT 1,2,3,4), eye closure (PVT 2,3), pupil diameter (PVT 1,2,3,4). These results are discussed and integrated in an ergonomic approach in order to analyze the links between objective performance, eye movements and subjective assessment of drowsiness. [less ▲]

Detailed reference viewed: 19 (4 ULg)
Peer Reviewed
See detailDetection of eye movements specific to drowsiness and their relation with subjective assessment of sleepiness in PVT: A cognitive ergonomic approach
Blavier, Adelaïde ULg; Montagnino, Cédric; Wertz, Jérôme ULg et al

Conference (2014, February 24)

Drowsiness is one of the major factors explaining accidents, particularly in traffic accidents but also in work situations with serious consequences (e.g. medicine). The drowsiness may be assessed by ... [more ▼]

Drowsiness is one of the major factors explaining accidents, particularly in traffic accidents but also in work situations with serious consequences (e.g. medicine). The drowsiness may be assessed by diverse measures that vary from physiological and unconscious data (e.g. EEG) to subjective and conscious evaluation. In their daily life, people are used to evaluate their drowsiness by subjective assessment and research observes a great inter-individual variation in this evaluation. Moreover, the subjective evaluation is dependent on the situation and the risk perceived by the person (e.g., new versus usual situations, simple versus complex environments, etc.). In this theoretical context, our purpose was to investigate the links between 1) objective performance (reaction time) measured by a psychomotor vigilance task (PVT), 2) data from eye movements and 3) subjective assessment of drowsiness (measured with Karolinska Sleepiness Scale, KSS). 12 persons, aged from 20 to 56, participated individually. They were asked to respect a 60% sleep deprivation during the night before the experiment and to not drink any energy drinks the day of the experiment. The experiment was conducted between 1 and 3 PM after a heavy lunch in order to increase the circadian effect. Moreover, the temperature of the room was 25° in order to increase drowsiness. Each participant was asked to perform 4 PVT. However, although 100% of participants performed the first two PVT, only 66% were able to perform the third PVT and only 33% of participants performed the fourth and last PVT. Our results showed an effect of time on objective performance, eye movements and subjective assessment of drowsiness in PVT 1 and 2: significant increase of reaction time, increase of eye closure and perclos, reduction of pupil diameter and increase of subjective drowsiness estimation (KSS). Only the frequency and duration of blinks remained constant across time. In PVT 3 and 4, data from eye movements and objective performance (reaction time) did not vary anymore across the time. Only the subjective estimation of drowsiness (KSS) continued to increase. Furthermore, subjective estimation of drowsiness was differently correlated with eye movements and objective performance across the PVT; it was significantly correlated with reaction time (PVT 1,2,3), blink frequency and duration (PVT 1,2,3), perclos (PVT 1,2,3,4), eye closure (PVT 2,3), pupil diameter (PVT 1,2,3,4). These results are discussed and integrated in an ergonomic approach in order to analyze the links between objective performance, eye movements and subjective assessment of drowsiness. [less ▲]

Detailed reference viewed: 29 (2 ULg)
Full Text
Peer Reviewed
See detailComparative evaluation of existing and new methods for correcting ocular artifacts in electroencephalographic recordings
Kirkove, Murielle ULg; François, Clémentine ULg; Verly, Jacques ULg

in Signal Processing (2014), 98(C), 102-120

EEG signals are often contaminated by ocular artifacts (OAs), in particular when they are recorded for a subject that is, in principle, awake, such as in a study of drowsiness. It is generally desirable ... [more ▼]

EEG signals are often contaminated by ocular artifacts (OAs), in particular when they are recorded for a subject that is, in principle, awake, such as in a study of drowsiness. It is generally desirable to detect and/or correct these OAs before interpreting the EEG signals. We have identified 11 existing methods for dealing with OAs. Their study allowed us to create 16 new methods. We performed a comparative performance evaluation of the resulting 27 distinct methods using a common set of data and a common set of metrics. The data was recorded during a driving task of about two hours in a driving simulator. This led to a ranking of all methods, with five emerging clear winners, comprising two existing methods and three new ones. [less ▲]

Detailed reference viewed: 28 (10 ULg)
Full Text
See detailDrowsiness monitoring for road safety
Wertz, Jérôme ULg; François, Clémentine ULg; Verly, Jacques ULg

Conference (2013, December 13)

Detailed reference viewed: 14 (1 ULg)
Full Text
Peer Reviewed
See detailPerformance evaluation of methods for correcting ocular artifacts in electroencephalographic (EEG) recordings
Kirkove, Murielle ULg; François, Clémentine ULg; Libotte, Aurélie et al

Conference (2013, February)

The presence of ocular artifacts (OA) due to eye movements and eye blinks is a major problem for the analysis of electroencephalographic (EEG) recordings in most applications. A large variety of methods ... [more ▼]

The presence of ocular artifacts (OA) due to eye movements and eye blinks is a major problem for the analysis of electroencephalographic (EEG) recordings in most applications. A large variety of methods (algorithms) exist for detecting or/and correcting OA’s. We identified the most promising methods, implemented them, and compared their performance for correctly detecting the presence of OA’s. These methods are based on signal processing “tools” that can be classified into three categories: wavelet transform, adaptive filtering, and blind source separation. We evaluated the methods using EEG signals recorded from three healthy persons subjected to a driving task in a driving simulator. We performed a thorough comparison of the methods in terms of the usual performances measures (sensitivity, specificity, and ROC curves), using our own manual scoring of the recordings as ground truth. Our results show that methods based on adaptive filtering such as LMS and RLS appear to be the best to successfully identify OA’s in EEG recordings. [less ▲]

Detailed reference viewed: 100 (40 ULg)
Full Text
See detailMonitoring de la somnolence
Wertz, Jérôme ULg; François, Clémentine ULg; Verly, Jacques ULg

Conference (2012, November 09)

Detailed reference viewed: 53 (21 ULg)
Full Text
See detailDrowsiness monitoring: a matter of life and death !
Wertz, Jérôme ULg; François, Clémentine ULg; Verly, Jacques ULg

Conference (2012, March 19)

Detailed reference viewed: 41 (20 ULg)