Using long short term memory and convolutional neural networks for driver drowsiness detection.

Fatigue negatively affects the safety and performance of drivers on the road. In fact, drowsiness and fatigue are the cause of a substantial number of motor vehicle accidents. Drowsiness among the drivers can be detected using variety of modalities, including electroencephalogram (EEG), eye movement, and vehicle driving dynamics. Among these EEG is highly accurate but very intrusive and cumbersome. On the other hand, vehicle driving dynamics are very easy to acquire but accuracy is not very high. Eye movement based approach is very attractive in terms of balance between these two extremes. However, eye movement based techniques normally require an eye tracking device which consists of high speed camera with sophisticated algorithm to extract eye movement related parameters such as blinking, eye closure, saccades, fixation etc. This makes eye tracking based drowsiness detection difficult to implement as a practical system, especially on an embedded platform. In this paper, authors propose to use eye images from camera directly without the need for expensive eye-tracking system. Here, eye related movements are captured by Recurrent Neural Network (RNN) to detect the drowsiness. Long Short Term Memory (LSTM) is a class of RNN which has several advantages over vanilla RNNs. In this work an array of LSTM cells are utilized to model the eye movements. Two types of LSTMs were employed: 1-D LSTM (R-LSTM) which is used as baseline and the convolutional LSTM (C-LSTM) which facilitates using 2-D images directly. Patches of size 48 × 48 around each eye were extracted from 38 subjects, participating in a simulated driving experiment. The state of vigilance among the subjects were independently assessed by power spectral analysis of multichannel electroencephalogram (EEG) signals, recorded simultaneously, and binary labels of alert and drowsy (baseline) were generated. Results show high efficacy of the proposed system. R-LSTM based approach resulted in accuracy around 82 % and C-LSTM based approach resulted in accuracy in the range of 95%-97%. Comparison is also provided with a recently published eye-tracking based approach, showing the proposed LSTM technique outperform with a wide margin.

Sustained vigilance is negatively affected by mild and acute sleep loss reflected by reduced capacity for decision making, motor preparation, and execution.

Study Objectives: The behavioral and cognitive consequences of severe sleep deprivation are well understood. Surprisingly, relatively little is known about the neural correlates of mild and acute sleep restriction on tasks that require sustained vigilance for prolonged periods of time during the day. Methods and Results: Event-related potential (ERP) paradigms can reveal insight into the neural correlates underlying visual processing and behavioral responding that is impaired with reduced alertness, as a consequence of sleep loss. Here, we investigated the impact of reduced vigilance following at-home mild sleep restriction to better understand the associated behavioral consequences and changes in information processing revealed by ERPs. As expected, vigilance was reduced (e.g. increased lapses and response slowing) that increased over the course of the experiment in the "sleep restricted" (5 hr sleep) compared with the "sleep-extension" (9 hr sleep) condition. Corresponding to these lapses, we found decreased positivity of visually evoked potentials in the Sleep Restriction vs. Sleep Extension condition emerging from 316 to 449 ms, maximal over parietal/occipital cortex. We also investigated electrophysiological signs of motor-related processing by comparing lateralized readiness potentials (LRPs) and found reduced positivity of LRPs in the Sleep Restriction vs. Sleep Extension condition at 70-40 ms before, and 115-158 ms after a response was made. Conclusions: These results suggest that even a single night of mild sleep restriction can negatively affect vigilance, reflected by reduced processing capacity for decision making, and dulls motor preparation and execution.

Semantic image retrieval in magnetic resonance brain volumes.

Practitioners in the area of neurology often need to retrieve multimodal magnetic resonance (MR) images of the brain to study disease progression and to correlate observations across multiple subjects. In this paper, a novel technique for retrieving 2-D MR images (slices) in 3-D brain volumes is proposed. Given a 2-D MR query slice, the technique identifies the 3-D volume among multiple subjects in the database, associates the query slice with a specific region of the brain, and retrieves the matching slice within this region in the identified volumes. The proposed technique is capable of retrieving an image in multimodal and noisy scenarios. In this study, support vector machines (SVM) are used for identifying 3-D MR volume and for performing semantic classification of the human brain into various semantic regions. In order to achieve reliable image retrieval performance in the presence of misalignments, an image registration-based retrieval framework is developed. The proposed retrieval technique is tested on various modalities. The test results reveal superior robustness performance with respect to accuracy, speed, and multimodality.

Adaboost and Support Vector Machines for White Matter Lesion Segmentation in MR Images.

The use of two powerful classification techniques (boosting and SVM) is explored for the segmentation of white-matter lesions in the MRI scans of human brain. Simple features are generated from Proton Density (PD) scans. Radial Basis Function (RBF) based Adaboost technique and Support Vector Machines (SVM) are employed for this task. The classifiers are trained on severe, moderate and mild cases. The segmentation is performed in T1 acquisition space rather than standard space (with more slices). Hence, the proposed approach requires less time for manual verification. The results indicate that the proposed approach can handle MR field inhomogeneities quite well and is completely independent from manual selection process so that it can be run under batch mode. Segmentation performance comparison with manual detection is also provided.