[IEEE 2011 6th International Conference on Broadband and Biomedical Communications (IB2Com) -...

Abstract—Electroencephalography can be utilized to detect

driver fatigue. One algorithm that shows promising results in detecting fatigue is the equation ((θ+α)/β). The current study observes the result of equation ((θ+α)/β) on 10 train drivers who were completing a 30-minute monotonous train driving experiment.

Index Terms—Electroencephalography, Train driver, fatigue monitor

I. INTRODUCTION

Train drivers and controllers (or dispatchers) are normally required to work on a 24-hour irregular and rotating shift [1]. Shiftwork may interrupt the normal sleep-wake circadian cycle, which can lead to fatigue and performance impairment [2-5]. Several studies have found that at least 50% of train drivers and traffic controllers were severely fatigued during night shifts, and about 25% of train drivers fell asleep while driving the train [6, 7].

Fatigue is a significant factor in industrial accidents, especially in the 24-hour industries with long hours, night shifts, or rotating shifts, such as in railway industries [8]. Several train accidents have been attributed to fatigue, such as the Waterfall train accident in New South Wales (NSW), Australia in January 2003 [8], and the coal train accident in Beresfield, NSW in October 1997 [9]. Train drivers often drive in a monotonous environment, where the scenery and the demands of the task vary very little from one day to another. Monotony and boredom are part of the job [6], and this often reduces the attention of drivers to signals while driving, and advances the driver to a fatigued state [10, 11].

The cost of train accidents not only affect those who are

Manuscript received August 18, 2011. The research was supported by an

Australian Research Council Linkage grant (LP0560886) and by SENSATION Integrated Project (FP6-507231) co-funded by the Sixth Framework Programme of the European Commission under the Information Society Technologies priority.

Budi Thomas Jap, BEng (Computer System), BA (International Studies), is currently completing his PhD at University of Technology, Sydney (UTS), Australia. (+612-9512-8291, email: [email protected]).

Peter Fischer, BEng (Electrical), Fellow of the Institute of Engineers, Australia. Director of Signal Network Technologies Pty Ltd. ([email protected]).

Sara Lal, (PhD, MAppSc, GCHE, DipLaw) is an academic supervisor at the University of Technology, Sydney. ([email protected])

directly involved in the accidents, such as the drivers and the passengers, but also the community and the general population [12]. For example, long-term psychological problems comprise some part of the cost of train accidents. After experiencing a train accident, approximately 16% of train drivers develop post-traumatic stress disorder, which can last for years [13-15]. Physical injury, such as restricted mobility in different parts of the body, disturbance of sight or hearing, and discomfort from scar or cartilage formations, also causes long-term problems for some of the victims [16]. The consequences of an accident to the victims can be long lasting and may affect them for a lifetime.

In order to avoid fatigue related train accidents, all parties involved in the railway industries, such as drivers, controllers, and company managers, need to cooperate and have the same perception and strategy towards fatigue prevention [17, 18]. Education can allow better understanding that everyone could be prone to driving fatigue, and need to take rests regularly on a long journey [17]. Although it is impossible to avoid night and early morning shifts, or irregularity in shift in a 24-hour railway industry, it is possible to adjust the shift lengths, the time between shifts and shift combinations in order to minimize fatigue [6]. However, it is not always possible to allow for such flexibility into the shift as the driver is compelled to work beyond the mental and physical capacity.

Several fatigue countermeasure techniques and devices have been researched and studied, and these may be useful to combat fatigue related train accidents. Automatic Train Protection (ATP) was developed to prevent trains passing signals at dangerous speeds, or failing to stop on terminating lines [19]. Others have developed fatigue countermeasure devices that detect eye closure, eye gaze, and blink rate measurement [20, 21]. Some facial feature detection techniques have also been developed to observe the changes in facial expressions, such as yawning, glazed look, and frequent nodding, which can be used to detect fatigue [22, 23]. However, fatigue may occur with or without obvious appearances of drowsy behaviours, such as yawning, eye-blink, and nodding, and micro-sleeps may have occurred long before any physical appearance of drowsy mannerisms [24].

Another possible fatigue countermeasure device that has been studied extensively is the electroencephalography (EEG) to measure the brain activity. Artaud et al. [25] found that EEG is one of the most reliable indicators of fatigue. The alert

Assessing a potential electroencephalography based algorithm during a monotonous train

driving task in train drivers

Budi T Jap, Peter Fischer, Sara Lal

Proceedings of the 6th International Conference on Broadband Communications & Biomedical Applications, November 21 - 24, 2011, Melbourne, Australia

139 IB2COM 2011

brain state has a mixture of alpha and beta activities [26]. An increase in alpha and theta activities accompanied by a decrease in beta activity can be interpreted as an indication of drowsiness [27, 28]. Jap et al. [29] have also shown significant decreases in beta activity as one gets fatigued during a monotonous train driving experiment.

Several studies have indicated that the use of combined EEG frequency bands may be a more reliable fatigue indicator [30-32]. The equation (θ+α)/β can be used as a reliable fatigue indicator, since it combines the theta and alpha activities to detect changes in the alertness level [30]. The equation provides a ratio between the slow wave (θ and α) activities and the fast wave (β) activity. As one gets fatigued, the output of the equation is expected to show a significant increase in activity [29].

Hence, the aim of this study was to examine the changes in the equation (θ+α)/β (identified as promising by Jap et al. [29, 33]) on 10 train drivers during a monotonous train driving task.

II. METHODS

A total of male train drivers, aged 21-65 years (mean: 44 ± 9.4 years), were recruited to participate in the monotonous train driving study. All participants provided informed consent prior to entering the study, and held a current Rail Safety Worker Certificate (Driver) from the Department of Transport, Australia.

Approval for this study has been obtained from the institute’s Human Research Ethics Committee (UTS HREC REF NO. 2006-176A). The study was conducted in a temperature-controlled laboratory. All participants were asked to refrain from consuming coffee, tea, or food approximately 4 hours prior to the study to minimise brain activity variation due to the effects food and beverage consumption [34]. Smoking was also restricted approximately 4 hours prior to the study, and participants were asked to refrain from alcohol consumption approximately 24 hours prior. Lifestyle Appraisal Questionnaire was used to establish selection criteria, that is, participants had to have no medical contraindication that could limit compliance, such as severe concomitant disease, alcoholism, drug abuse, and psychological or intellectual problems [35].

A. Selection of the 10 participants

The Fatigue Likert scale, which was aimed to identify the immediate fatigue level, and was administered prior to and after the driving study, was utilised to identify participants that could be included in this case study. The range of score for the Fatigue Likert scale was from 1 (alert) to 4 (extremely fatigued).

Participants that were included in the current case study scored a 1 (alert) or 2 (slightly fatigued) prior to the monotonous driving session, and scored a 3 (moderately fatigued) or 4 (extremely fatigued) after the monotonous driving session. These participants would provide an observable brain activity transition from the alert state to the

fatigued state.

B. Processing of EEG Recordings

The EEG recording during the 30-minute monotonous train driving session was sectioned into 30 one-minute epochs. These epochs were then subjected to Fast Fourier Transform (FFT) to derive the four frequency components (delta (δ) (0–4 Hz), theta (θ) (4–8 Hz), alpha (α) (8–13 Hz), and beta (β) (13–35 Hz)) [36]. After transforming the EEG recording into frequency domain, the EEG spectra were divided into 1-minute sections, and all epochs in the 1-minute section were averaged to obtain a value for each 1-minute section. This resulted in 30 one-minute average values for the 30-minute of monotonous train driving session.

The first 5 minutes of the monotonous driving session was calculated as the alert baseline and used as a comparison baseline against the EEG activity for the remaining driving session.

C. Selection of Algorithm

Previous studies by Eoh et al. [30] and Jap et al. [31, 37] have also shown that a combination of EEG frequency bands could be used to detect the transition from the alert state to the fatigue state. A combination of theta, alpha, and beta frequency bands in the equation ((θ+α)/β) has been shown to increase as the driver starts getting fatigued. This equation provides a ratio between slow wave EEG activities (theta and alpha) and fast wave EEG activity (beta). Slow wave EEG activities has been shown to increase during fatigue [30, 38], while fast wave activity decreases during fatigue [39, 40].

Therefore, the algorithm that will be used for this simulation study is the equation ((θ+α)/β).

III. RESULT

A total of 10 train driver volunteers, aged 36-47 years (mean: 42±3.6 years), were included in this case study. The total average driving time for the train drivers were 32±2.9 minutes. Studies have indicated that monotonous driving for 20-30 minutes could lower alertness level and induce driver fatigue [41, 42]. One volunteer (volunteer 7) was extremely fatigued during the monotonous driving session, and the study was stopped after 24 minutes of monotonous driving. Table 1 presents the age, driving duration, and the pre- and post-study Fatigue Likert score of the 10 train drivers that were included in this case study.

From Table 1, the volunteers can be separated into three groups based on the pre- and post-study Likert scale. The first group is for those volunteers (volunteers 1, 4, 5, and 10) who were in an alert state (Likert scale 1) at the beginning of the study, and were in a moderately fatigued state (Likert scale 3) at the end of the study. The second group consists of volunteers (volunteers 2, 6, and 9) who were alert (Likert scale 1) prior to the study, and were extremely fatigued (Likert scale 4) after the study. The last group consists of volunteers (volunteer 3, 7, and 8) who were slightly fatigued (Likert scale 2) at the beginning of the study and were extremely fatigued


140 IB2COM 2011

(Likert scale 4) after the study. The combination of theta (θ), alpha (α), and beta (β)

activities in the equation (θ+α)/β was expected to increase as the individual gets fatigued [30]. An increase of theta activity and a decrease of beta activity have been associated with deterioration of driving performance due to fatigue [29, 30, 32, 43, 44].

A. Volunteer group 1

The first group of volunteers consists of those who reported to be alert (Likert scale 1) at the beginning of the driving session and were moderately fatigued (Likert scale 3) at the end of the driving session (volunteers 1, 4, 5, and 10). The result of the (θ+α)/β activities for this group showed a 100% increase from the 5-minute baseline activity (Figure 1). Volunteers 1 and 10 had a steady increase of activity throughout the driving session, while volunteers 4 and 5 had large fluctuations of (θ+α)/β activity throughout the driving session, which averaged to approximately a 100% increase from the 5-minute baseline activity at the end of the driving session.

Figure 1 Volunteers 1, 4, 5, and 10: frontal activity; showing percentage change from a 5-minute baseline average; θ = theta; α = alpha; β = beta

B. Volunteer group 2

The group of volunteers, who were alert (Likert scale 1) at the start of the driving session and were extremely fatigued

(Likert scale 4) at the end of the driving session, showed a higher increase of (θ+α)/β activity as a result. At the end of the driving session, volunteers 2 and 6 showed that (θ+α)/β activity was 130%-160% higher than the 5-minute baseline (Figure 2). Volunteer 9 showed a greater increase of 200% in (θ+α)/β activity from the baseline activity.

Figure 2 Volunteers 2, 6, and 9: frontal activity; showing percentage change from a 5-minute baseline average; θ = theta; α = alpha; β = beta

C. Volunteer group 3

The third group of volunteers consisted of those who were slightly fatigued (Likert scale 2) at the beginning of the driving session and extremely fatigued (Likert scale 4) after the driving session. This group of volunteers showed a similar trend as the first group that reported to be alert (Likert scale 1) prior to the driving session and was moderately fatigued (Likert scale 3) at the end of the driving session. Volunteers 3 and 8 had a steady increase of activity which was approximately 100% higher than baseline (Figure 3). However, volunteer 7, who was extremely tired throughout the driving session, had a greater increase of approximately 200% higher than the baseline activity after 15 minutes of driving monotonously.

Figure 3 Volunteers 3, 7, and 8: frontal activity; showing percentage change from a 5-minute baseline average; θ = theta; α = alpha; β = beta

TABLE I AGE, DRIVING TIME, AND PRE- AND POST-STUDY FATIGUE LIKERT SCORE

# AGE TOTAL DRIVING TIME

(MINUTES) PRE-STUDY

LIKERT POST-STUDY

LIKERT

1 47 32 1 3 2 42 32 1 4 3 44 32 2 4 4 43 31 1 3 5 40 32 1 3 6 36 32 1 4 7 38 24 2 4 8 46 31 2 4 9 38 34 1 4

10 41 35 1 3


141 IB2COM 2011

IV. DISCUSSION

Electroencephalography (EEG) can be used to detect fatigue and has been found to be quite reliable [25]. A good test and retest reliability and high reproducibility of EEG activity has also been found during two episodes of monotonous driving sessions for the delta and theta bands [45], while others have reported acceptable reliability for alpha and beta activities [46, 47].

Beta activity recorded from an EEG during a monotonous driving task has been shown to decrease significantly when an individual is fatigued [29, 30]. According to Belyavin & Wright [39], the most sensitive indicator of fatigue is the significant decrease in beta activity, although others have argued that an increase in alpha activity is the most reliable indicator of reduced vigilance [48]. A decrease in beta activity has been associated to worsening performance and reduced vigilance [30, 38, 39].

The current study have investigated the percentage change in activity for the output of the equation (θ+α)/β proposed by Brookhuis & Waard [32] and Eoh et al. [30]. Equation (θ+α)/β combines both theta and alpha powers to detect changes in the alertness level [30]. Our studies and those of others have found that theta activity increases as one becomes fatigued [30, 31, 38]. The activity resulted from the output of equation (θ+α)/β has been shown to increase as the individual gets fatigued, as a result of a decrease in beta activity and an increase in theta [30, 37]. Eoh et al. [30] believed that this equation, which combined alpha and theta powers, could be a basis for a more reliable indicator of fatigue, especially “during the repetitive phase transition between wakefulness and micro sleep” [30].

The result of the current study for the equation (θ+α)/β shows an increasing trend as the individuals get fatigued. At the end of the driving session, the resultant activity for the equation increased by approximately 100% when compared to a 5-minute baseline average for the individuals who reported to moderately fatigued after the driving session. This increase was greater (between 150%-200%) for those who reported to be extremely fatigued at the end of the driving session. For the volunteers who were slightly fatigued at the start of the driving session, the percentage increase of the results of equation (θ+α)/β was approximately 100%, since the baseline average activity might already be higher due to a pre-existing slightly fatigued state. The result of the current study is similar to the findings that other studies have reported in the literature [30, 31, 37].

This study has shown that monitoring EEG activity changes would be useful as a fatigue countermeasure device. A 100% increase of the activity output from the equation (θ+α)/β would be a sign that the individual is moderately fatigued. A combined effort of these two EEG activity changes would confirm whether or not the individual is moderately fatigued. However, when the individual has commenced driving while in a fatigued state, the EEG baseline may not always reflect the correct EEG alert activities. This may produce an incorrect result, since the individual may already be extremely fatigued when the 100% increase of output from equation (θ+α)/β

activity occurred subsequently. Some studies proposed a fatigue detection technique

through the use of neural network or wavelet transform to classify alertness and drowsiness [26, 38, 49]. While others proposed techniques using Independent Component Analysis (ICA) algorithm to remove EEG artefacts as well as estimating drowsiness [4, 50]. These techniques may be combined with detecting the change in EEG activities to improve the detection rate. As Heitmann et al. [51] has observed that no single technology or technique is reliable enough for detecting driver fatigue, there is the need to combine different technologies or techniques together in order to increase reliability and reduce false alarms. The drawbacks of one technique may be compensated with the strength of the others.

V. CONCLUSION

The current study has investigated the changes of the resultant activity as an output of equation (θ+α)/β, proposed by Brookhuis & Waard [32] and Eoh et al. [30], on ten train drivers. There are three different groups of drivers that were included in this case study. The first group consisted of those who were alert at the start of the driving session, and were moderately fatigued at the end of the driving session. This group of drivers showed about 100% increase in activity from equation (θ+α)/β at the end of the driving session. The second group reported to be alert prior to the driving session, and were extremely fatigued at the end of the driving session. The result of this group was approximately 150%-200% increase in activity derived from equation (θ+α)/β at the end of the session. The last group of drivers began the monotonous driving session in a slightly fatigued state, and completed the driving session in an extremely fatigued state. An increase of approximately 100% in activity from equation (θ+α)/β was recorded for this last group.

Future studies may need to look at combining the above algorithm ((θ+α)/β) with other algorithms, such as the use of neural network, wavelet transform, or ICA algorithm, to detect changes in EEG activities and to increase reliability and reduce false alarms, as suggested by Heitmann et al. [51].

ACKNOWLEDGMENT

The research was supported by an ARC Linkage grant Australia (LP0560886) and by SENSATION Integrated Project (FP6-507231) co-funded by the Sixth Framework Programme of the European Commission under the Information Society Technologies priority.

REFERENCES [1] J. Dorrian, S. D. Baulk, and D. Dawson, “Work hours, workload,

sleep and fatigue in Australian Rail Industry employees,” Applied Ergonomics, vol. 42, no. 2, pp. 202-209, 2011.

[2] D. F. Dinges, “An overview of sleepiness and accidents,” Journal of Sleep Research, vol. 4, no. Supplement 2, pp. 4-14, 1995.

[3] P. Thiffault, and J. Bergeron, “Monotony of road environment and driver fatigue: a simulator study,” Accident Analysis & Prevention, vol. 35, no. 3, pp. 381-391, 2003.


142 IB2COM 2011

[4] C.-T. Lin, R.-C. Wu, S.-F. Liang et al., “EEG-based drowsiness estimation for safety driving using independent component analysis,” IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 52, no. 12, pp. 2726- 2738, 2005.

[5] S. A. Ferguson, N. Lamond, K. Kandelaars et al., “The Impact of Short, Irregular Sleep Opportunities at Sea on the Alertness of Marine Pilots Working Extended Hours,” Chronobiology International: The Journal of Biological & Medical Rhythm Research, vol. 25, no. 2/3, pp. 399, 2008.

[6] M. Härmä, M. Sallinen, R. Ranta et al., “The effect of an irregular shift system on sleepiness at work in train drivers and railway traffic controllers,” Journal of Sleep Research, vol. 11, no. 2, pp. 141-151, 2002.

[7] A. Austin, and P. D. Drummond, “Work problems associated with suburban train driving,” Applied Ergonomics, vol. 17, no. 2, pp. 111-116, 1986.

[8] National Transport Commission (NTC), Fatigue Management Within The Rail Industry: Review of Regulatory Approach, National Transport Commission (NTC), Melbourne, Australia, 2004.

[9] Australian Transport Safety Bureau (ATSB), Coal Train Collision Beresfield NSW 23 October 1997, NSW Department of Transport, Transport Safety Bureau, Canberra, Australia, 1998.

[10] G. D. Edkins, and C. M. Pollock, “The influence of sustained attention on Railway accidents,” Accident Analysis & Prevention, vol. 29, no. 4, pp. 533-539, 1997.

[11] G. Hildebrandt, W. Rohmert, and J. Rutenfranz, “12 and 24 hour rhythms in error frequency of locomotive drivers and the influence of tiredness,” International Journal of Chronobiology, vol. 2, pp. 175-180, 1974.

[12] Bureau of Transport and Regional Economics (BTRE), Rail accident costs in Australia, 108, Department of Transport and Regional Services, Commonwealth of Australia, Canberra, 2002.

[13] R. Farmer, T. Tranah, I. O'Donnell et al., “Railway suicide: the psychological effects on drivers,” Psychological Medicine, vol. 22, no. 2, pp. 407-414, 1992.

[14] A. Jabez, “Back in control,” Nursing Times, vol. 89, no. 39, pp. 46-47, 1993.

[15] S. Karlehagen, U. F. Malt, H. Hoff et al., “The effect of major railway accidents on the psychological health of train drivers--II. A longitudinal study of the one-year outcome after the accident,” Journal of Psychosomatic Research, vol. 37, no. 8, pp. 807-817, 1993.

[16] A.-L. Andersson, L.-O. Dahlbäck, and P. Allebeck, “Psychosocial consequences of traffic accidents: a two year follow-up,” Scandinavian Journal of Social Medicine, vol. 22, no. 4, pp. 299-302, 1994.

[17] P. H. Gander, N. S. Marshall, W. Bolger et al., “An evaluation of driver training as a fatigue countermeasure,” Transportation Research Part F: Traffic Psychology and Behaviour, vol. 8, no. 1, pp. 47-58, 2005.

[18] P. K. Arnold, and L. R. Hartley, “Policies and practices of transport companies that promote or hinder the management of driver fatigue,” Transportation Research Part F: Traffic Psychology and Behaviour, vol. 4, no. 1, pp. 1-17, 2001.

[19] A. W. Evans, “The economics of automatic train protection in Britain,” Transport Policy, vol. 3, no. 3, pp. 105-110, 1996.

[20] W. W. Wierwille, L. A. Ellsworth, S. S. Wreggit et al., Research on vehicle-based driver status/performance monitoring: development, validation, and refinement of algorithms for detection of driver drowsiness, National Highway Traffic Safety Administration Final Report: DOT HS 808 247, 1994.

[21] G. Longhurst, "Understanding Driver Visual Behaviour," http://www.seeingmachines.com/pdfs/media_coverage/2002/2002_TTI_p92.pdf, [14 May 2009, 2002].

[22] H. Gu, Q. Ji, and Z. Zhu, "Active facial tracking for fatigue detection."

[23] Q. Ji, and X. Yang, “Real-Time Eye, Gaze, and Face Pose Tracking for Monitoring Driver Vigilance,” Real-Time Imaging, vol. 8, pp. 357–377, 2002.

[24] J. A. Horne, and L. A. Reyner, “Vehicle accidents related to sleep: a review,” Occupational and Environmental Medicine, vol. 56, no. 5, pp. 289-294, 1999.

[25] P. Artaud, S. Planque, C. Lavergne et al., "An on-board system for detecting lapses of alertness in car driving."

[26] M. K. Kiymik, M. Akin, and A. Subasi, “Automatic recognition of alertness level by using wavelet transform and artificial neural network,” Journal of Neuroscience Methods, vol. 139, no. 2, pp. 231-240, 2004.

[27] S. K. L. Lal, and A. Craig, “A critical review of the psychophysiology of driver fatigue,” Biological Psychology, vol. 55, no. 3, pp. 173-194, 2001.

[28] J. Santamaria, and K. H. Chiappa, “The EEG of Drowsiness in Normal Adults,” Journal of Clinical Neurophysiology, vol. 4, no. 4, pp. 327-382, 1987.

[29] B. T. Jap, S. Lal, and P. Fischer, “Comparing combinations of EEG activity in train drivers during monotonous driving,” Expert Systems with Applications, vol. 38, no. 1, pp. 996-1003, 2011.

[30] H. J. Eoh, M. K. Chung, and S.-H. Kim, “Electroencephalographic study of drowsiness in simulated driving with sleep deprivation,” International Journal of Industrial Ergonomics, vol. 35, no. 4, pp. 307-320, 2005.

[31] B. T. Jap, S. Lal, P. Fischer et al., “Using EEG spectral components to assess algorithms for detecting fatigue,” Expert Systems with Applications, vol. 36, no. 2, Part 1, pp. 2352-2359, 2009.

[32] K. A. Brookhuis, and D. Waard, “The use of psychophysiology to assess driver status,” Ergonomics, vol. 39, no. 9, pp. 1099-1110, 1993.

[33] S. K. L. Lal, R. Henderson, S. Waters et al., “Differences in EEG hemispheric activity during fatigue,” in 2nd international conference of the EU 6th framework integrated project SENSATION, Chania, Greece, 2007.

[34] J. L. Kenemans, and M. M. Lorist, “Caffeine and selective visual processing,” Pharmacology Biochemistry and Behavior, vol. 52, no. 3, pp. 461-471, 1995.

[35] A. Craig, K. Hancock, and M. Craig, “The Lifestyle Appraisal Questionnaire: A Comprehensive Assessment of Health and Stress,” Psychology and Health, vol. 11, pp. 331-343, 1996.

[36] A. J. Rowan, and E. Tolunsky, Primer of EEG: With a Mini-Atlas, USA: Elsevier Science, 2003.

[37] B. T. Jap, S. Lal, P. Fischer et al., "Using Spectral Analysis to Extract Frequency Components from Electroencephalography: Application for Fatigue Countermeasure in Train Drivers."

[38] A. Subasi, “Automatic recognition of alertness level from EEG by using neural network and wavelet coefficients,” Expert Systems with Applications, vol. 28, no. 4, pp. 701-711, 2005.

[39] A. Belyavin, and N. A. Wright, “Changes in electrical activity of the brain with vigilance,” Electroencephalography and Clinical Neurophysiology, vol. 66, no. 2, pp. 137-144, 1987.

[40] S. K. L. Lal, and A. Craig, “Driver Fatigue: Electroencephalography and Psychological Assessment,” Psychophysiology, vol. 39, pp. 313-321, 2002.

[41] M. Gillberg, G. Kecklund, and T. Åkerstedt, “Sleepiness and performance of professional drivers in a truck simulator-comparisons between day and night driving,” Journal of Sleep Research, vol. 5, no. 1, pp. 12-15, 1996.

[42] A. Craig, "Vigilance: theories and laboratory studies," Hours of Work: Temporal Factors in Work Scheduling, S. Folkard and T. Monk, eds., Chichester: Wiley, 1985.

[43] D. R. Davies, “Skin conductance, alpha activity and vigilance,” American Journal of Psychology, vol. 78, no. 2, pp. 304-306, 1965.

[44] B. T. Jap, S. Lal, S. Waters et al., "Comparison of different algorithms based on EEG spectral analysis during continuous driving."

[45] S. K. L. Lal, and A. Craig, “Reproducibility of the spectral components of the electroencephalogram during driver fatigue,” International Journal of Psychophysiology, vol. 55, no. 2, pp. 137-143, 2005.

[46] V. E. Pollock, L. S. Schneider, and S. A. Lyness, “Reliability of topographic quantitative EEG amplitude in healthy late-middle-aged and elderly subjects,” Electroencephalography and Clinical Neurophysiology, vol. 79, no. 1, pp. 20-26, 2002.

[47] A. J. Tomarken, R. J. Davidson, R. E. Wheeler et al., “Psychometric properties of resting anterior EEG asymmetry: temporal stability and internal consistency,” Psychophysiology, vol. 29, no. 5, pp. 576– 592, 1992.

[48] L. Torsvall, and T. Åkerstedt, “Sleepiness on the job: continuously measured EEG changes in train drivers,” Electroencephalography and Clinical Neurophysiology, vol. 66, no. 6, pp. 502-511, 1987.


143 IB2COM 2011

[49] T.-P. Jung, S. Makeig, M. Stensmo et al., “Estimating alertness from the EEG power spectrum,” IEEE Transactions on Bioimedical Engineering, vol. 44, no. 1, pp. 60-69, 1997.

[50] J.-P. Lanquart, M. Dumontb, and P. Linkowski, “QRS artifact elimination on full night sleep EEG,” Medical Engineering & Physics, vol. 28, no. 2, pp. 156-165, 2006.

[51] A. Heitmann, R. Guttkuhn, A. Aguirre et al., "Technologies for the Monitoring and Prevention of Driver Fatigue."

Budi Jap, BEng (Computer System), BA (International Studies), is currently completing his PhD at University of Technology, Sydney (UTS), Australia. The main research area is in driver fatigue and finding the means to detect driver fatigue from different physiological signals, such as brain or heart activities. Peter Fischer, BEng (Electrical), Fellow of the Institute of Engineers, Australia. Director of Fischer Industries Pty Ltd, and Director of Signal Network Technologies Pty Ltd. Researching and developing railway safety and driver vigilance systems for over 40 years. Expert in physiological and physical monitoring fields. Sara Lal, PhD, is an academic supervisor at the University of Technology, Sydney. Some of Dr Lal's areas of research are neuroscience, cardiovascular, sleep disorder, cognitive function, transportation safety, and sensor and technology development. Dr Lal has attracted various competitive grants and has published a book, book chapters, refereed journals and conference papers.


144 IB2COM 2011

[IEEE 2011 6th International Conference on Broadband and Biomedical Communications (IB2Com) -...

Documents

Transcript of [IEEE 2011 6th International Conference on Broadband and Biomedical Communications (IB2Com) -...