다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

I have developed "tennis elbow" from lugging this book around the past four weeks, but it is worth the pain, the effort, and the aspirin. It is also worth the (relatively speaking) bargain price. Including appendixes, this book contains 894 pages of text. The entire panorama of the neural sciences is surveyed and examined, and it is comprehensive in its scope, from genomes to social behaviors. The editors explicitly state that the book is designed as "an introductory text for students of biology, behavior, and medicine," but it is hard to imagine any audience, interested in any fragment of neuroscience at any level of sophistication, that would not enjoy this book. The editors have done a masterful job of weaving together the biologic, the behavioral, and the clinical sciences into a single tapestry in which everyone from the molecular biologist to the practicing psychiatrist can find and appreciate his or

Principles of Neural Science

技術解説 IEEE Computer

Emotion recognition in human-computer interaction

For physical activity recognition, smartphone sensors, such as an accelerometer and a gyroscope, are being utilized in many research studies. So far, particularly, the accelerometer has been extensively studied. In a few recent studies, a combination of a gyroscope, a magnetometer (in a supporting role) and an accelerometer (in a lead role) has been used with the aim to improve the recognition performance. How and when are various motion sensors, which are available on a smartphone, best used for better recognition performance, either individually or in combination? This is yet to be explored. In order to investigate this question, in this paper, we explore how these various motion sensors behave in different situations in the activity recognition process. For this purpose, we designed a data collection experiment where ten participants performed seven different activities carrying smart phones at different positions. Based on the analysis of this data set, we show that these sensors, except the magnetometer, are each capable of taking the lead roles individually, depending on the type of activity being recognized, the body position, the used data features and the classification method employed (personalized or generalized). We also show that their combination only improves the overall recognition performance when their individual performances are not very high, so that there is room for performance improvement. We have made our data set and our data collection application publicly available, thereby making our experiments reproducible.

/pdf/fusion-of-smartphone-motion-sensors-for-physical-activity-4xdnlyv3ro.pdf

Fusion of smartphone motion sensors for physical activity recognition.

Template matching is an alternative to perform speech recognition. However, the template matching encountered problems due to speaking rate variability, in which there exist timing differences between the two utterances. Speech has a constantly changing signal, thus it is almost impossible to get the same signal for two same utterances. The problem of time differences can be solved through DTW algorithm: warping the template with the test utterance based on their similarities. So, DTW algorithm actually is a procedure, which combines both warping and distance measurement. DTW is considered as one effective method in speech pattern recognition, however the bad side of this method is that it requires a long processing time plus large storage capacity, especially for real time recognitions.

Dynamic time warping

Electroencephalography (EEG) is a non-invasive process that record and capture brain signal. Raw EEG signal contaminated with a lot of noise such as power line interference, muscle movement (electromyography artifacts) and eye blinking (electrooculography artifacts). Removing this types of noise can produce a clean signal. The noise that contaminated with EEG signal will affect the actual result during analyzing stage. This paper aims to shows the comparison when EEG signal filter with 4th order Butterworth band pass and stationary wavelet transform (SWT). Two parameters are used to compare the effect of filter on EEG signal, that are mean square error (MSE) and peak-to-noise ratio (PSNR). This types of parameter was use because it able to compare the quantitative value for two signals. The result shows that the stationary wavelet transform is more effective in removing the noise without losing the original information.

Butterworth Bandpass and Stationary Wavelet Transform Filter Comparison for Electroencephalography Signal

Electromyography (EMG) is widely used in various fields to investigate the muscular activities. Since EMG signals contain a wealth of information about muscle functions, there are many approaches in analyzing the EMG signals. It is important to know the features that can be extracting from the EMG signal. The ideal feature is important for the achievement in EMG analysis. Hence, the objective of this paper is to evaluate the features extraction of time domain from the EMG signal. The experiment was setup according to surface electromyography for noninvasive assessment of muscle (SENIAM). The recorded data was analyzed in time domain to get the features. Based on the analysis, three features have been considered based on statistical features. The features was then been evaluate by getting the percentage error of each feature. The less percentage error determines the ideal feature. The results shows that the extracted features of the EMG signals in time domain can be implement in signal classification. These findings could be integrated to design a signal classification based on the features extraction.

/pdf/features-extraction-of-electromyography-signals-in-time-1sz99srmg0.pdf

Features extraction of electromyography signals in time domain on biceps brachii muscle

Today, machine learning and deep learning have paved the way for vital and critical applications such as abnormal detection. Despite the modernity of transfer learning, it has proved to be one of the crucial inventions in the field of deep learning because of its promising results. For the purpose of this study, transfer learning is utilized to extract human motion features from RGB video frames to improve detection accuracy. A convolutional neural network (CNN) based on Visual Geometry Group network 19 (VGGNet-19) pre-trained model is used to extract descriptive features. Next, the feature vector is passed into Binary Support Vector Machine classifier (BSVM) to construct a binary-SVM model. The performance of the proposed framework is evaluated by three parameters: accuracy, area under the curve, and equal error rate. Experiments performed on two different datasets comprising highly different context abnormalities accomplished an accuracy of 97.44% and an area under the curve (AUC) of 0.9795 for University of Minnesota (UMN) dataset and accomplished an accuracy of 86.69% and an AUC of 0.7987 for University of California, San Diego Pedistrain1 (UCSD-PED1) dataset. Moreover, the performance of the pre-trained network VGGNet-19 with handcrafted feature descriptors and with other CNN pre-trained networks, respectively, has been investigated in this study for abnormal behavior detection. The results demonstrated that VGGNet-19 has better performance than histogram of oriented gradients, background subtraction, and optical flow. In addition, the VGGNet-19 shows higher detection accuracy than other pre-trained networks: GoogleNet, ResNet50, AlexNet, and VGGNet-16.

Transfer Deep Learning Along With Binary Support Vector Machine for Abnormal Behavior Detection

Inefficient energy use has been a major issue globally. In Malaysia, the statistics reveal that residential energy consumption has been on a steady increase due to the growing population as well as a lack of awareness within households regarding proper energy utilization that causes significant amount of energy wastage. The emergence of Internet-of-Things (IoT) is a consequence and convergence of several key technologies such as real-time analytics, machine learning, sensors and embedded systems. The application of intelligence in IoT, known as Cognitive IoT (CIoT), will enable decision making based on historical data, and automatically train, learn and troubleshoot future issues. This project proposes a smart energy monitoring system for home appliances incorporating CIoT which consists of three parts. Firstly, a Raspberry Pi-based smart plug serving as the gateway, that is able to read current data from individual home appliances, load the trained model from training server and test the verified data using the model. Secondly, Google Colab as the training server will be used to store the training data set and building the Tensorflow-based Long Short-term Memory (LSTM) model. This recurrent neural network model will forecast electricity bill and notify users if abnormal energy consumption of individual home appliances is detected. Thirdly, a dashboard using Matplotlib library where users may monitor the real-time energy consumption. The completed LSTM model demonstrates a high accuracy of more than 80%, where a low mean squared error with train score = 0.1798 and test score = 0.1229 is determined. Furthermore, the R2 test shows a high level of goodness of fit with train score = 0.844 and test score = 0.835.

Rubita Sudirman

Papers

Dynamic time warping

Butterworth Bandpass and Stationary Wavelet Transform Filter Comparison for Electroencephalography Signal

Features extraction of electromyography signals in time domain on biceps brachii muscle

Transfer Deep Learning Along With Binary Support Vector Machine for Abnormal Behavior Detection

Machine Learning for Smart Energy Monitoring of Home Appliances Using IoT