An Effective Hybrid Deep Learning Model for Single-Channel EEG-Based Subject-Independent Drowsiness Recognition

Nowadays, road accidents pose a severe risk in cases of sleep disorders. We proposed a novel hybrid deep-learning model for detecting drowsiness to address this issue. The proposed model combines the strengths of discrete wavelet long short-term memory (DWLSTM) and convolutional neural networks (CNN) models to classify single-channel electroencephalogram (EEG) signals. Baseline models such as support vector machine (SVM), linear discriminant analysis (LDA), back propagation neural networks (BPNN), CNN, and CNN merged with LSTM (CNN+LSTM) did not fully utilize the time sequence information. Our proposed model incorporates a majority voting between LSTM layers integrated with discrete wavelet transform (DWT) and the CNN model fed with spectrograms as images. The features extracted from sub-bands generated by DWT can provide more informative & discriminating than using the raw EEG signal. Similarly, spectrogram images fed to CNN learn the specific patterns and features with different levels of drowsiness. Furthermore, the proposed model outperformed state-of-the-art deep learning techniques and conventional baseline methods, achieving an average accuracy of 74.62%, 77.76% (using rounding, F1-score maximization approach respectively for generating labels) on 11 subjects for leave-one-out subject method. It achieved high accuracy while maintaining relatively shorter training and testing times, making it more desirable for quicker drowsiness detection. The performance metrics (accuracy, precision, recall, F1-score) are evaluated after 100 randomized tests along with a 95% confidence interval for classification. Additionally, we validated the mean accuracies from five types of wavelet families, including daubechis, symlet, bi-orthogonal, coiflets, and haar, merged with LSTM layers.

A Motion Estimation based Algorithm for Encoding Time Reduction in HEVC

High Efficiency Video Coding (HEVC) is a video compression standard that offers 50% more efficiency at the expense of high encoding time contrasted with the H.264 Advanced Video Coding (AVC) standard. The encoding time must be reduced to satisfy the needs of real-time applications. This paper has proposed the Multi- Level Resolution Vertical Subsampling (MLRVS) algorithm to reduce the encoding time. The vertical subsampling minimizes the number of Sum of Absolute Difference (SAD) computations during the motion estimation process. The complexity reduction algorithm is also used for fast coding the coefficients of the quantised block using a flag decision. Two distinct search patterns are suggested: New Cross Diamond Diamond (NCDD) and New Cross Diamond Hexagonal (NCDH) search patterns, which reduce the time needed to locate the motion vectors. In this paper, the MLRVS algorithm with NCDD and MLRVS algorithm with NCDH search patterns are simulated separately and analyzed. The results show that the encoding time of the encoder is decreased by 55% with MLRVS algorithm using NCDD search pattern and 56% with MLRVS using NCDH search pattern compared to HM16.5 with Test Zone (TZ) search algorithm. These results are achieved with a slight increase in bit rate and negligible deterioration in output video quality.

Analysis on improving operational efficiency of construction project site

Operational efficiency directly reflects the profitability of a project. It is for benefit of a construction site. The current research paper focus on high operational efficiency by reducing waste, better layout planning, sound administration, business practices an, review of existing facilities etc. To improve the operational efficiency will enhance the profit by adding the value to the project in the form of cost reduction in internal transportation, utilization of existing facilities effectively. The study adopted the Computerized Relative Allocation of Facilities Techniques (CRAFT) and computerized relationship planning (CORELAP) algorithm to analyze the operational efficiency.

Sexy males and sexless females: the origin of triploid apomicts

Apomixis and polyploidy are closely associated in angiosperms, but the evolutionary reason for this association is unknown. Taraxacum officinale, the common dandelion, exists both as diploid sexuals and triploid apomicts. Here, in the context of T. officinale, we provide a model of the evolution of triploid apomicts from diploid sexuals. We posit an apomictic allele that arrests female meiosis in diploids, so that the plant produces diploid egg cells that can develop without fertilization, but haploid pollen. We propose occasional fertilization of diploid egg cells by haploid pollen, resulting in triploid apomicts that produce triploid egg cells but largely nonfunctional pollen. The irreversibility of this process renders diploid partial apomicts evolutionarily short-lived, and results in fixation of triploid apomicts except when they suffer extreme selective disadvantages. Our model can account for the high genetic diversity found in T. officinale triploid populations, because recombinant haploid pollen produced by diploids allows the apomictic allele to spread onto many genetic backgrounds. This leads to multiple clonal lineages in the newly apomictic population, and thereby alleviates some of the usual pitfalls of asexual reproduction.

A new sphingolipid from the gorgonianjunceella junceaof the Indian ocean

A new sphingolipid, (2S,3R,4E)-1,3-dihydroxy-2-[(nonadecanoyl) amino]-octadec-4-ene (1) along with ubiquitous batyl alcohol is isolated from the gorgonian, Junceella juncea Pallas (Gorgonaceae) of the Indian Ocean. The structure of 1 is determined on the basis of 1H- and 13C-NMR, COSY, FABMS, and GC-EIMS experiments.

New sphingosines from a gorgonian, Pseudopterogorgia australiensis Ridley, of the Indian Ocean

Two new sphingosines, (2S,3R)-2-(docosanoyl amino)nonadecane-1,3-diol (1) and (2S,3S,4R)-2-[(2'R)-2'-hydroxynonadecanoylamino]nonadecane-1,3,4-triol (3), along with the known (2S,3R,4E)-2-(heptadecanoylamino)octadec-4-ene-1,3-diol, have been isolated from Pseudopterogorgia australiensis, of the Indian Ocean. The structures were deduced from spectral and chemical methods. Compounds 1-3 showed moderate antibacterial activity.

Oxytocin receptor couples to the 80 kDa Gh alpha family protein in human myometrium

One of the primary functions of the oxytocin receptor is to modulate intracellular calcium levels in myometrium. The oxytocin receptor has been purified and cloned. Although it has been suggested that oxytocin receptor couples with a GTP-binding regulatory protein (G-protein), the identity of this G-protein remains unclear. To elucidate the mechanism of oxytocin receptor signalling, we used the oxytocin-receptor-G-protein ternary complex preparation from human myometrium, and evaluated oxytocin-mediated activation of [35S]guanosine 5'-[gamma-thio]triphosphate ([35S]GTP[S]) binding and [alpha-32P]GTP photoaffinity labelling to a G-protein. Binding of [35S]GTP[S] and the intensity of the [alpha-32P]GTP photoaffinity labelled protein resulting from activation of the oxytocin receptor were significantly attenuated by the selective oxytocin antagonist, desGlyNH2d(CH2)5[Tyr(Me)2,Thr4]OVT. Furthermore, the molecular mass of the specific GTP-binding protein was approximately 80 kDa; homologous with the Gh alpha family, the new class of GTP-binding proteins first identified in rat liver that couples to the alpha 1B-adrenoceptor. Consistent with these observations, in co-immunoprecipitation and co-immunoadsorption of the oxytocin receptor in the ternary complex preparation by anti-Gh7 alpha antibody, the Gh alpha family protein tightly coupled to the oxytocin receptor. These findings demonstrate that oxytocin receptor couples with approximately 80 kDa Gh alpha in signal mediation.