Proposal for Prevention of Drowsy Driving-Related Motor Vehicle Accidents in Patients With Obstructive Sleep Apnea

Obstructive sleep apnea (OSA) is known to be associated with various health concerns, including sleepiness, fatigue, cognitive dysfunction, diminished quality of life, hypertension, cardiovascular diseases, and stroke. OSA-induced sleepiness at the wheel reduces vigilance and driving performance, which significantly increase the risk of motor vehicle accidents. Sleepiness-induced motor vehicle accidents are characterized by high morbidity and mortality. OSA is a well-established significant risk factor for drowsy driving-related motor vehicle accidents, which can be prevented through appropriate treatment. However, currently no clinical guidelines or regulations are available for evaluation or management of the risk of motor vehicle accidents in patients with OSA in Korea. In this review, we discuss the risk of motor vehicle accidents in patients with OSA, the effects of positive airway pressure therapy as a preventive measure to reduce this risk, and the published recommendations for OSA in other countries with regard to fitness to drive. We propose recommendations for screening, evaluation, and treatment of OSA with regard to the risk of motor vehicle accidents, which would serve as useful practical guidelines for sleep specialists in clinical practice. Further research is warranted to establish optimal strategies for effective improvements in OSA-related traffic safety.

Proposal for Preventing Motor Vehicle Accident Related to Drowsy Driving in Patients With Narcolepsy

Narcolepsy is a chronic sleep disorder characterized by irresistible sleep attacks, hypersomnolence, cataplexy (sudden loss of muscle tone provoked by emotion), and sleep paralysis. Individuals with narcolepsy are at a high risk of experiencing sleepiness while driving leading to road traffic accidents. To prevent such accidents, some countries have regulations for commercial and noncommercial drivers with narcolepsy. Evaluating sleepiness is essential. Therefore, several subjective reports and objective tests were used to predict the possibility of car crashes or near-misses. Brain stimulants are effective in treating narcolepsy and can reduce daytime sleepiness in these patients. However, no guideline has been established for the driving safety of patients with narcolepsy in Korea. The Korean Sleep Research Society has prepared this proposal for preventing motor vehicle accidents caused by drowsy driving in patients with narcolepsy.

Real-Time V2V Communication With a Machine Learning-Based System for Detecting Drowsiness of Drivers

This article applies machine learning to detect whether a driver is drowsy and alert the driver. The drowsiness of a driver can lead to accidents resulting in severe physical injuries, including deaths, and significant economic losses. Driver fatigue resulting from sleep deprivation causes major accidents on today's roads. In 2010, nearly 24 million vehicles were involved in traffic accidents in the U.S., which resulted in more than 33,000 deaths and over 3.9 million injuries, according to the U.S. NHTSA. A significant percentage of traffic accidents can be attributed to drowsy driving. It is therefore imperative that an efficient technique is designed and implemented to detect drowsiness as soon as the driver feels drowsy and to alert and wake up the driver and thereby preventing accidents. The authors apply machine learning to detect eye closures along with yawning of a driver to optimize the system. This paper also implements DSRC to connect vehicles and create an ad hoc vehicular network on the road. When the system detects that a driver is drowsy, drivers of other nearby vehicles are alerted.

P017 Am I sleepy? – Subjective Sleepiness and Drowsy Driving: A systematic review and meta-analysis

Introduction Driving impairment due to sleep loss is a major contributor to motor vehicle crashes resulting in severe injury or fatalities. Ideally, drivers should be aware of their sleepiness and cease driving to reduce risk of a crash. However, there is little consensus on how accurately drivers can identify sleepiness, and how this relates to subsequent driving impairment. To examine whether drivers are aware of their sleepiness, we systematically reviewed the literature. Methods The research question for this review was “are drivers aware of sleepiness while driving, and to what extent does subjective sleepiness accurately reflect driving impairment?”. Our search strategy led to thirty-four simulated/naturalistic driving studies for review. We then extracted the relevant data. Correlational data were examined using meta-analysis, while predictive data were assessed via narrative review. Results Results showed that drivers were aware of sleepiness, and this was associated with both driving impairment and physiological drowsiness. Overall, subjective sleepiness was more strongly correlated (a) with ocular and EEG-based outcomes (rweighted = .70 and .73, respectively, p<.001), rather than lane position and speed outcomes (rweighted = .46 and .49, respectively, p<.001); (b) under simulated driving conditions compared to naturalistic drives; and (c) when the Karolinska Sleepiness Scale was used to measure subjective sleepiness. Lastly, high levels of sleepiness significantly predicted crash events and lane deviations. Discussion This review presents evidence that drivers are aware of sleepiness when driving, and suggests that interventions such as stopping driving when feeling ‘sleepy’ may significantly reduce crash risk.

Accident Analysis and Identification of Black Spots on the Motorways in Pakistan - a Reliability Analysis Approach

The objective of this study is to explore the contributory factors responsible for road accidents and identifies the black spots on the three motorways; M1 (Peshawar-Islamabad), M2 (Islamabad-Lahore), and M3 (Pindi BhattianFaisalabad) in Pakistan. Five years’ road accident data was obtained from the National Highways and Motorway Police (NHMP), Pakistan. The database of this study included six hundred road accidents on a total of 574 kilometers long routes of M1, M2 and M3. The reliability analysis approach was used to locate black spot locations on each motorway. For the visualization and mapping of black spots on each motorway, a Geographic Information System (GIS) was used. The results explored that vehicle condition was the significant contributory factor responsible for the maximum number of road accidents on M1 and M3, while for M2, it was drowsy driving. It is also found that a maximum number of road accidents on M2 and M3 occurred in late-night, while for M1, it was day timing. Furthermore, road accidents were relatively higher in May-July and December on M1 and M2, which shows that extreme weather influences the occurrence of road accidents. On the contrary, no substantial variation of road accidents was examined for M3 month-wise. Finally, black spots on each motorway were located and their georeferenced coordinates were presented for future use. As a result, precautionary measures and provisions are suggested for concerned authorities to mitigate road safety problems.

Drowsiness Detection for Motorized Vehicles Using Machine Learning

Driver drowsiness is one of the major causes for most of the accidents in the world. Detecting the driver's eye tiredness is the easiest way for measuring the drowsiness of the driver. The advent of high-speed motorized vehicles drowsy driving accidents has claimed the lives of millions of people across the globe. To avoid such accidents, proposes a Machine Learning based system drowsiness system for motorized vehicles with alarm and Web Push Notifications to notify the driver before any accident occurs. The driver's face is captured by a real-time camera system, and the eye borders are detected by a pre-trained machine learning model from the real-time video stream. Then each eye is represented by 6 – coordinates (x, y) starting from the left corner of the eye and then working clockwise around the eye. The EAR (Ear Aspect Ratio) is calculated across 20 consecutive frames, and if it falls below a certain threshold, it sounds an alarm and sends the details of the nearest coffee shop to your mobile device via a Web Push Notification. When the alarm is activated, it also displays a list of nearby coffee shops to help the driver stay awake. Keywords: Machine Learning, SVM, MOR, EAR

Going Beyond Position Statements: One Hospital’s Successful Initiative to Implement Napping for Night Shift Nurses

Background: Sleepiness during the night shift is associated with errors, accidents, injuries, and drowsy driving. Despite scientific evidence that supports brief naps to reduce sleepiness, and guidance documents from policy organizations, napping has not been widely implemented. Methods: An initiative to translate scientific evidence about napping was implemented in one hospital over one year. The initiative included garnering leadership support and resources, building a translation team, evaluating the evidence, responding to operational concerns, developing an implementation strategy, and then implementing and evaluating the results. Night shift nurses were surveyed pre and post nap implementation for drowsy driving, sleepiness, and work and coworker relationships. Qualitative data documented the nurses’ perceptions about napping. Findings: Three-fourths of the units that were eligible to nap successfully implemented and sustained napping. Most nurses felt refreshed by a brief nap and felt safer on the drive home, but one-fourth worried about or had sleep inertia symptoms. Drowsy driving remained unacceptably high. Conclusion: The initiative was successfully implemented on most nursing units. The mixed reaction to napping, and the unfavorable drowsy driving outcome point to the need for additional interventions to reduce sleepiness.

An Efficient Approach for Detecting Driver Drowsiness Based on Deep Learning

Drowsy driving is one of the common causes of road accidents resulting in injuries, even death, and significant economic losses to drivers, road users, families, and society. There have been many studies carried out in an attempt to detect drowsiness for alert systems. However, a majority of the studies focused on determining eyelid and mouth movements, which have revealed many limitations for drowsiness detection. Besides, physiological measures-based studies may not be feasible in practice because the measuring devices are often not available on vehicles and often uncomfortable for drivers. In this research, we therefore propose two efficient methods with three scenarios for doze alert systems. The former applies facial landmarks to detect blinks and yawns based on appropriate thresholds for each driver. The latter uses deep learning techniques with two adaptive deep neural networks based on MobileNet-V2 and ResNet-50V2. The second method analyzes the videos and detects driver’s activities in every frame to learn all features automatically. We leverage the advantage of the transfer learning technique to train the proposed networks on our training dataset. This solves the problem of limited training datasets, provides fast training time, and keeps the advantage of the deep neural networks. Experiments were conducted to test the effectiveness of our methods compared with other methods. Empirical results demonstrate that the proposed method using deep learning techniques can achieve a high accuracy of 97% . This study provides meaningful solutions in practice to prevent unfortunate automobile accidents caused by drowsiness.