Driver sleepiness contributes to a large amount of all road traffic crashes. Developing an objective measurement of driver sleepiness in order to prevent eventual traffic accidents is desirable. The aim of this master thesis was to investigate if deep learning can be used to provide a driver sleepiness classification from brain activity signals obtained by electroencephalography (EEG). The intention was to study the classification performance when using different representations of the input data and to examine how various deep neural network architectures and class weighting during training affect the classification.  The data was collected from 12 experiments, where 269 participants (1187 driving sessions) were driving either on real roads or in a moving-base driving simulator, while electrophysiological data was recorded. Several deep neural network architectures were developed, depending on the representation of the input data.  Regardless of which data representation that was used as input to the network, the datawas divided into three datasets: Training 60%, validation 20% and test 20%. The data from each participant, with associated driving sessions, were randomly assigned to the different datasets according to the given percentage, which resulted in a subject-independent sleepiness detection. The output was in the form of continuous regression further rounded to the closest integer and divided into five classes according to Karolinska Sleepiness Scale (KSS = 1-5, 6, 7, 8, 9). The best performance was obtained with a convolutional neural network (CNN) combined with Long Short-Term Memory (LSTM) architecture, with time series data as input. This gave an accuracy of 41.44%, a mean absolute error of 0.94 and a macro F1-score of 0.37. Overall, the models with time series data showed better classification results compared to those with time-frequency data. Class weighting, giving all classes inverse proportional weight to their appearance, compensated slightly for class imbalance, but all networks had in general difficulties with generalizing to new data.

Bakgrund: Nattarbete utgör en hälsorisk ur flera aspekter och har flera konsekvenser liknande de som ohälsosamma matvanor har. Bland lastbilschaufförer förekommer långa körningar där hälsosamma matvanor inte främjas. När arbetet sker under natten förskjuts sömnen och enligt dygnsrytmen infaller en naturlig trötthet vilket kan orsaka problem i form av olyckor. Dygnsrytmen styr även hunger och ämnesomsättning. Syfte: Att undersöka arbetsmåltider och det möjliga sambandet med upplevd trötthet hos lastbilschaufförer under både dagkörningar och nattkörningar. Metod: En webbaserad enkät delades ut på ett företag samt i en Facebookgrupp och besvarades av 43 lastbilschaufförer. Resultat: Resultaten visade att det var kaffe, vatten, smörgåsar och sötsaker som intogs i högst utsträckning. Majoriteten av måltider intogs i lastbilen under körning och under nattkörningen intogs måltider främst vid kl. 20 samt 02. För att motverka trötthet intogs koffein. Samband visade att den upplevda tröttheten ökade när intaget av mat och dryck i lastbilen under körning ökade. Samband visade också att den upplevda tröttheten ökade när intag av mat och dryck vid kl. 00-02 minskade. Ytterligare samband visade att den upplevda tröttheten ökade när möjligheten att själv kunna bestämma tid för mat och dryck minskade. Slutsats: Chaufförers möjligheter att lämna lastbilen samt att själv välja när de vill äta och dricka är relevant att förbättra då det finns ett samband mellan detta och den trötthet chaufförerna upplever under sina körningar, både under dag och natt. Arbetsgivare har möjlighet att utforma arbetsorganisationen så att förbättringar kan underlättas. / Aim: To investigate work meals and the possible correlation to sleepiness among truck drivers during both day driving and night driving. Method: A web-based survey was distributed at a company and in a Facebook group. 43 truck drivers responded. Results: The results showed that the most common intake was coffee, water, sandwiches, and sweets. Most of the foods and drinks were consumed inside the truck, while driving. During the night drive foods and drinks were consumed mostly at 8 pm and 2 am. Caffeine was used to counteract sleepiness. Self-reported sleepiness increased as eating or drinking inside the truck, while driving, increased. Sleepiness also increased as eating or drinking at 12 pm-2 am decreased. Another association showed that sleepiness increased as having the option to choose when to eat or drink decreased.  Conclusion: Because of the associations found, it is highly relevant to improve truck driver’s opportunities to leave the truck and to have better options in choosing when to eat or drink. Employers have the possibility to make organisational changes to make improvements easier.

Driver sleepiness is a cause for crashes and it is estimated that 3.9 to 33 % of all crashes might be related to sleepiness at the wheel. It is desirable to get an objective measurement of driver sleepiness for reduced sensitivity to subjective variations. Using deep learning for classification of driver sleepiness could be a step toward this objective. In this master thesis, deep learning was used for investigating classification of electrophysiological data, electroencephalogram (EEG) and electrooculogram (EOG), from drivers into levels of sleepiness. The EOG reflects eye position and EEG reflects brain activity. Initially, the intention was to include electrocardiogram (ECG), which reflects heart activity, in the research but this data were later excluded. Both raw time series data and data transformed into time-frequency domain representations were fed into the developed neural networks and for comparison manually extracted features were used in a shallow neural network architecture. Investigation of using EOG and EEG data separately as input was performed as well as a combination as input. The data were labeled using the Karolinska Sleepiness Scale, and the scale was divided into two labels "fatigue" and "alert" for binary classification or in five labels for comparison of classification and regression. The effect of example length was investigated using 150 seconds, 60 seconds and 30 seconds data. Different variations of the main network architecture were used depending on the data representation and the best result was given when using a combination of a convolutional neural network (CNN) and a long short-term memory (LSTM) network with time distributed 150 seconds EOG data as input. The accuracy was in this case 80.4 % and the majority of both alert and fatigue epochs were classified correctly with 85.7 % and 66.7 % respectively. Using the optimal threshold from the created receiver operating characteristics (ROC) curve resulted in a more balanced classifier with 76.3 % correctly classified alert examples and 79.2 % correctly classified fatigue examples. The results from the EEG data, both in terms of accuracy and distribution of correctly classified examples, were shown to be less promising compared to EOG data. Combining EOG and EEG signals was shown to slightly increase the proportion of correctly classified fatigue examples. However, more promising results were obtained when balancing the classifier for solely EOG signals. The overall result from this project shows that there are patterns in the data connected to sleepiness that the neural network can find which makes further work on applying deep learning to the area of driver sleepiness interesting.

Commercial vehicle drivers are required to maintain Department Of Transportation medical certification which entails a Commercial Driver Medical Examination (CDME) and optimally leads to a two-year certification. The examination must be performed by a licensed "medical examiner" administered by a variety of health care providers including physicians, advanced registered nurse practitioners, physician assistants and doctors of chiropractic. Unfavorable findings in the examination can yield either a shortened medical certification period or denial of certification. Sleep disorders including sleep apnea are assessed by a single question located in the health history portion of the CDME form which is filled-out by the examinee. A positive response to this single item often prompts the medical examiner to further supplement this question using a subjective questionnaire, such as the Epworth Sleepiness Scale. This particular questionnaire generates a total score based on the examinee's subjective responses to eight items regarding the propensity to doze-off or fall asleep in different scenarios, thus indicating daytime sleepiness. Commercial drivers depend on the medical certification for their livelihood and it is hypothesized that subjective responses regarding daytime sleepiness are distorted in an effort to attain optimal DOT certification.

No description available.

Approximately 1.35 million people die each year in car accidents and it is the most common cause of death for people aged 5-29. Because of this it is of large interest to be able to detect when a driver enters a sleepy state and to be able to alert the driver. This thesis aims to develop and evaluate a video-based system for driver sleepiness classification, and also to investigate personalised classification. The dataset used in this work consisted of face videos of 89 drivers captured using an IR camera. All drivers drove on a highway in real traffic in four different driving sessions, two at night time while being sleep deprived and two at day time while being alert. The drivers estimated their sleepiness level every 5th minute according to the 9-point Karolinska Sleepiness Scale (KSS).  The proposed driver sleepiness detection system consists of two steps. First, facial features are extracted from videos. Facial features corresponding to 60 seconds of data were used as input to a 1D-convolutional network. The network architecture consisted of five convolution blocks follower by a fully connected (dense) layer and an output layer with linear activation. The KSS-ratings were used as target values. For classification, the output was rounded to the closest KSS-level. KSS-levels 1-4 were merged into a single class, giving a total of 6 classes. The hyperparameters of the developed models were set by finding the best performing model during a grid search. Personalised models were developed by leaving out data from one of the drivers during training, and then finetuning the model parameters of the last dense layer by training on three of the driving sessions from the left out driver and evaluating on the fourth session. On average, the generic models of the developed architecture obtained an accuracy of 22%, a mean absolute error (MAE) of 1.49 and macro-F1 of 0.16. For the personalised models the mean results were an accuracy of 21%, MAE of 1.50 and macro-F1 of 0.14. The generic model showed signs of bias towards the majority classes (low KSS-levels) despite using weighting to compensate for class imbalance. Further, the results did not improve when personalising the model to compensate for individual differences. Future work should investigate if a larger and more balanced dataset is the only remedy, or if algorithm-level methods can be used to get around the problem.

Bakgrund: Restless legs syndrome (RLS) är en underbehandlad sjukdom som påverkar välbefinnande och hälsa. Sömnsituationen hos personer med RLS är komplex och påverkas av symtom och varierande omständigheter. Förmågan att ta sig an de utmaningar som kommer med symtom på kvällen och natten erfars stressande och tidigare forskning tyder på att fler erfar symtom än antal personer som får diagnos, vilket leder till onödigt lidande.  Syfte: Syftet var att utforska och beskriva hur personer med RLS erfar sin sömnsituation.  Metod: En kvalitativ intervjustudie med manifest innehållsanalys genomfördes av 25 personer med RLS bosatta i olika delar av Sverige. KASAM utgör den teoretiska ramen i studien. Resultat: Stress över att känna symtom på kvällen och natten samt svårigheter att sova erfors. Rutiner och distraktion var viktiga för att kunna sova. Oro över att inte hitta symtomlindrande behandling återberättades av informanterna, som ofta själva sökte efter fungerande medicinsk behandling. Ständig trötthet erfors och kunde leda till psykiska konsekvenser men även svårigheter med sociala kontakter och på arbetet. Slutsats: Det finns stora kunskapsluckor om RLS och dess påverkan på sömnen, därför behövs mer forskning inom området för att distriktssköterskan ska kunna hjälpa personer med RLS att få en bättre sömnkvalitet och ökat välbefinnande. / Background: RLS is an undertreated disease that affects well-being and health. For people with RLS, sleeping is a complex siuation and affected by symptoms in varying forms. Coping with symptoms that come at night is described as stressful, and previous research suggests that more people than diagnosed experience symptoms, which leads to unneccessary suffering.  Purpose: To explore and describe how people with RLS experience their sleeping situation. Method: A qualitative interview study with manifest content analysis was conducted by 25 people with RLS living in Sweden. KASAM forms the theoretical framework of the study. Result: Stress over feeling symptoms in the evening and at night as well as difficulty sleeping. Routines and distraction were important to being able to sleep. Concerns about not finding symptom-relieving treatment was recounted by the informants, who often searched for effective medical treatment themselves. Constant fatigue was experienced and could lead to psychological consequences but also difficulties with social contacts and at work. Conclusion: There are large gaps in knowledge about RLS and its impact on sleep, therefore more research is needed in the area so that the district nurse can help people with RLS get a better quality of sleep and increased well-being.

No description available.

<p>At the same time as many urban economies are developing into 24-hour societies it is becoming increasingly popular amongst shift workers to compress their working hours. This is done by working longer shifts (>8h) and/or restricting free time (<16h) in between shifts – the main reasons are to gain longer bouts of free time and extra free weekends. However, there is a limited knowledge of the effects of such arrangements on sleep and wakefulness. Thus, the main purposes of the present thesis were to evaluate the effects of long working hours (in the form of 12h shifts) and short recovery periods. Another aim was to evaluate possible mechanisms that could suggest why some individuals develop problems with shift work while others do not.</p><p>We used a combination of methods - sleep diaries, wake diaries, blood samples and objective measures of sleep and cognitive performance - across whole or large parts of shift schedules to evaluate acute effects of particularly demanding working periods, as well as the total effects of a shift cycle. Study I evaluated the effect of changing from an 8h- to a 12h-shift system. Study II evaluated the effects of long shifts in a shift schedule with both 8h- and 12h-shifts. Study III evaluated the effects of several consecutive short recovery periods (8-9h of recovery) and whether satisfaction with ones’ work hours was associated to problems with sleep and sleepiness. Study IV evaluated whether endocrinological markers of catabolic (cortisol) and anabolic (testosterone) activity changed across a shift sequence and whether satisfaction were related to them. Study V was a laboratory simulation of the effects of a short recovery period (4h of sleep) and whether a short nap could counteract any detrimental effects.</p><p>There was no convincing evidence for 12h shifts inducing more problems with sleep and sleepiness than 8h shifts. With regard to recovery time between shifts, the shortest recovery times (only 8h) seriously shortened sleep duration and increased sleepiness, while 12h of recovery (between two consecutive 12h shifts) was judged as having no or limited effects on acute measures. The problems with the shortest recovery periods were worse in a schedule with several consecutive shifts and less pronounced in a schedule with few consecutive shifts. With regard to individual differences, it was found that subjects being dissatisfied with their working hours were vulnerable to short recovery periods, which was evident by less sufficient sleep and an accumulation of sleepiness across work periods with limited recovery time. Interestingly, these problems disappeared when they were allowed to recover after the work period. In addition, dissatisfied male shift workers had lower testosterone levels at the end of work periods, indicating disturbed anabolic activity. The simulated quick return supported that curtailed sleep affected sleepiness and performance and that a short nap could counteract these effects temporarily.</p><p>It is concluded that long shifts (up to 12h) may be acceptable, whereas short recovery time (8h or less) is not. Most of the problems with short recovery periods were related to short sleep and sleepiness, and there is, clearly, a subgroup of workers that suffer more from this than others. It is argued that insufficient sleep and low testosterone levels (in males) might be key factors for developing shift intolerance, mainly by reducing the capacity to recover from shift work.</p>

At the same time as many urban economies are developing into 24-hour societies it is becoming increasingly popular amongst shift workers to compress their working hours. This is done by working longer shifts (&gt;8h) and/or restricting free time (&lt;16h) in between shifts – the main reasons are to gain longer bouts of free time and extra free weekends. However, there is a limited knowledge of the effects of such arrangements on sleep and wakefulness. Thus, the main purposes of the present thesis were to evaluate the effects of long working hours (in the form of 12h shifts) and short recovery periods. Another aim was to evaluate possible mechanisms that could suggest why some individuals develop problems with shift work while others do not. We used a combination of methods - sleep diaries, wake diaries, blood samples and objective measures of sleep and cognitive performance - across whole or large parts of shift schedules to evaluate acute effects of particularly demanding working periods, as well as the total effects of a shift cycle. Study I evaluated the effect of changing from an 8h- to a 12h-shift system. Study II evaluated the effects of long shifts in a shift schedule with both 8h- and 12h-shifts. Study III evaluated the effects of several consecutive short recovery periods (8-9h of recovery) and whether satisfaction with ones’ work hours was associated to problems with sleep and sleepiness. Study IV evaluated whether endocrinological markers of catabolic (cortisol) and anabolic (testosterone) activity changed across a shift sequence and whether satisfaction were related to them. Study V was a laboratory simulation of the effects of a short recovery period (4h of sleep) and whether a short nap could counteract any detrimental effects. There was no convincing evidence for 12h shifts inducing more problems with sleep and sleepiness than 8h shifts. With regard to recovery time between shifts, the shortest recovery times (only 8h) seriously shortened sleep duration and increased sleepiness, while 12h of recovery (between two consecutive 12h shifts) was judged as having no or limited effects on acute measures. The problems with the shortest recovery periods were worse in a schedule with several consecutive shifts and less pronounced in a schedule with few consecutive shifts. With regard to individual differences, it was found that subjects being dissatisfied with their working hours were vulnerable to short recovery periods, which was evident by less sufficient sleep and an accumulation of sleepiness across work periods with limited recovery time. Interestingly, these problems disappeared when they were allowed to recover after the work period. In addition, dissatisfied male shift workers had lower testosterone levels at the end of work periods, indicating disturbed anabolic activity. The simulated quick return supported that curtailed sleep affected sleepiness and performance and that a short nap could counteract these effects temporarily. It is concluded that long shifts (up to 12h) may be acceptable, whereas short recovery time (8h or less) is not. Most of the problems with short recovery periods were related to short sleep and sleepiness, and there is, clearly, a subgroup of workers that suffer more from this than others. It is argued that insufficient sleep and low testosterone levels (in males) might be key factors for developing shift intolerance, mainly by reducing the capacity to recover from shift work.