Hybrid Convolutional Neural Network and Long Short-Term Memory Approach for Facial Expression Recognition

Facial Expression Recognition(FER)has been an importantfield of research for several decades.Extraction of emotional characteristics is crucial to FERs,but is complex to process as they have significant intra-class variances.Facial characteristics have not been completely explored in static pictures.Previous studies used Convolution Neural Networks(CNNs)based on transfer learning and hyperparameter optimizations for static facial emotional recognitions.Particle Swarm Optimizations(PSOs)have also been used for tuning hyperparameters.However,these methods achieve about 92 percent in terms of accuracy.The existing algorithms have issues with FER accuracy and precision.Hence,the overall FER performance is degraded significantly.To address this issue,this work proposes a combination of CNNs and Long Short-Term Memories(LSTMs)called the HCNN-LSTMs(Hybrid CNNs and LSTMs)approach for FERs.The work is evaluated on the benchmark dataset,Facial Expression Recog Image Ver(FERC).Viola-Jones(VJ)algorithms recognize faces from preprocessed images followed by HCNN-LSTMs feature extractions and FER classifications.Further,the success rate of Deep Learning Techniques(DLTs)has increased with hyperparameter tunings like epochs,batch sizes,initial learning rates,regularization parameters,shuffling types,and momentum.This proposed work uses Improved Weight based Whale Optimization Algorithms(IWWOAs)to select near-optimal settings for these parameters using bestfitness values.The experi-mentalfindings demonstrated that the proposed HCNN-LSTMs system outper-forms the existing methods.

Enoxaparin dosing errors in the emergency department

BACKGROUND:The study aimed to determine the frequency of enoxaparin dosing errors for patients who had a measured emergency department(ED)weight compared to those who did not have a measured ED weight,and to determine if demographic variables(e.g.,weight,height,age,Englishspeaking,race)impact the likelihood of receiving an inappropriate dose.METHODS:This is a retrospective,electronic chart review of patients who received a dose of enoxaparin in the ED between January 1,2008 and July 1,2013.We identified all patients>18 years who received a dose of enoxaparin while in the ED,were admitted,and had at least one inpatient weight within the first four days of hospitalization.Patients were excluded if they received enoxaparin for prophylaxis or a dose of more than 1.25mg/kg.RESULTS:A total of 1,944 patients were included.Patients were more likely to experience an error if they did not have a measured ED weight.Over-doses of>10mg were more likely to occur in patients without a measured ED weight.Patients with no documented ED weight or with a staffestimated ED weight were more likely to experience a dosing error than those with a patient-stated weight.Patients were more likely to experience an error if their first inpatient weight was more than 96kg,they were more than 175-cm tall,or were English speaking.CONCLUSION:Dosing errors are more likely to occur when patients are not weighed in the ED.Modifications to current workflows to incorporate weighing those patients who receive weightdosed medications may be warranted.

Prediction of Abidjan Groundwater Quality Using Machine Learning Approaches: An Exploratory Study

Continuous groundwater quality monitoring poses significant challenges affecting the environment and public health. Groundwater in Abidjan, specifically from the Continental Terminal (CT), is the primary supply source. Therefore, ensuring safe drinking water and environmental protection requires a thorough evaluation and surveillance of this resource. Our present research evaluates the quality of the CT groundwater in Abidjan using the water quality index (WQI) based on the analytical hierarchy process (AHP). This study also explores the application of machine learning predictions as a time-efficient and cost-effective approach for groundwater resource management. Therefore, three Machine Learning regression algorithms (Ridge, Lasso, and Gradient Boosting (GB)) were executed and compared. The AHP-based WQI results classified 98.98% of samples as “good” water quality, while 0.68% and 0.34% of samples were respectively categorized as “excellent” and “poor” water. Afterward, the prediction performance evaluation highlighted that the GB outperformed the other models with the highest accuracy and consistency (MSE = 0.097, RMSE = 0.300, r = 0.766, rs = 0.757, and τ = 0.804). In contrast, the Lasso model recorded the lowest prediction accuracy, with an MSE of 148.921, an RMSE of 6.828, and consistency parameters of r = 0.397, rs = 0.079, and τ = 0.082. Gradient Boosting regression effectively learns nonlinear events and interactions by iteratively fitting new models to errors of previous models, enabling a more realistic groundwater quality prediction. This study provides a novel perspective for improving groundwater quality management in Abidjan, promoting real-time tracking and risk mitigations.

Attack Detection for Spoofed Synchrophasor Measurements Using Segmentation Network

Synchrophasor measurements are essential to realtime situational awareness of the smart grid but vulnerable to cyber-attacks during the process of transmission and invocation.To ensure data security and mitigate the impact of spoofed synchrophasor measurements,this work proposes a novel object detection method using a Weight-based One-dimensional Convolutional Segmentation Network(WOCSN)with the ability of attack behavior identification and time localization.In WOCSN,automatic data feature extraction can be achieved by onedimensional convolution from the input signal,thereby reducing the impact of handcrafted features.A weight loss function is designed to distribute the contribution for normal and attack signals.Then,attack time is located via the proposed binary method based on pixel segmentation.Furthermore,the actual synchrophasor data collected from four locations are used for the performance evaluation of the WOCSN.Finally,combined with designed evaluation metrics,the time localization ability of WOCSN is validated in the scenarios of composite attacks with different spoofed intensities and time-sensitivities.