2017 International Conference on Frontiers of Information Technology (FIT)最新文献

筛选

英文中文

EEG Brain Connectivity Analysis to Detect Driver Drowsiness Using Coherence 利用相干性检测驾驶员困倦的脑电图脑连接分析

2017 International Conference on Frontiers of Information Technology (FIT) Pub Date : 2017-07-02 DOI: 10.1109/FIT.2017.00027

M. Awais, N. Badruddin, M. Drieberg

{"title":"EEG Brain Connectivity Analysis to Detect Driver Drowsiness Using Coherence","authors":"M. Awais, N. Badruddin, M. Drieberg","doi":"10.1109/FIT.2017.00027","DOIUrl":"https://doi.org/10.1109/FIT.2017.00027","url":null,"abstract":"Drowsiness at the wheel is one of the major contributing factors towards road accidents. Therefore, efforts have been made to detect driver drowsiness using electroencephalogram (EEG). The use of EEG as a possible driver drowsiness indicator is commonly accepted. However, in this paper, we have studied brain connectivity measure instead of the traditional spectral power measures. For this purpose, the EEG coherence analysis is performed to examine the functional connectivity between various brain regions during the transitional phase, i.e., from alert state to drowsy state. Data collection is performed in a simulator based environment. Twenty-two healthy subjects voluntarily participated in the study after providing their consent. All possible combinations of inter- and intra-hemispheric coherences are analyzed. Because of the unavailability of common gold standard, video recordings are captured during the experiment to mark the drowsy state. To verify the statistical significance of the proposed features, paired t-test is performed. The analysis revealed significant differences (p0.05) in inter- and intra-hemispheric coherences (brain connectivity analysis) between alert and drowsy state, which shows the potential of coherence analysis in detection drowsiness.","PeriodicalId":107273,"journal":{"name":"2017 International Conference on Frontiers of Information Technology (FIT)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122381146","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 8

Deep and Self-Taught Learning for Protein Accessible Surface Area Prediction 蛋白质可接近表面积预测的深度自学

2017 International Conference on Frontiers of Information Technology (FIT) Pub Date : 2017-01-18 DOI: 10.1109/FIT.2017.00054

F. Hassan, F. Minhas

{"title":"Deep and Self-Taught Learning for Protein Accessible Surface Area Prediction","authors":"F. Hassan, F. Minhas","doi":"10.1109/FIT.2017.00054","DOIUrl":"https://doi.org/10.1109/FIT.2017.00054","url":null,"abstract":"ASA captures the degree of burial or surface accessibility of a protein residue. It is a very important indicator of the behavior of amino acids within a protein as well. It can be used to find protein interactions, interfaces, folding states, etc. Calculation of the ASA requires the presence of the structure of the protein. However, structure determination for proteins is expensive and requires significant technical effort. As a consequence, the prediction of ASA is a very important and fundamental problem in Bioinformatics and Proteomics. In this work, we have investigated self-taught machine learning methods along with deep neural network to predict the residue level accessible surface area (ASA) of a protein. We have found that deep learning neural networks can predict the ASA of the residues in a protein accurately. Furthermore, the proposed deep learning based method does not require the use of computationally demanding features such as the position specific scoring matrix (PSSM) which have been used in previous works. A simple Blosum62 matrix based position dependent representation of amino acids in a sequence window gives comparable performance. This is particularly attractive for proteome wide prediction of ASA. We have used various self-taught learning schemes for obtaining an optimal feature representation from unlabeled data. These include a sparse and regularized autoencoder neural network and a dictionary based learning scheme. We have used unlabeled data from the protein universe in an attempt to improve the feature representation. We have also evaluated the performance of a stochastic gradient based predictor of accessible surface area for different feature representations.","PeriodicalId":107273,"journal":{"name":"2017 International Conference on Frontiers of Information Technology (FIT)","volume":"144 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123275368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

首页上一页