2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)最新文献

{"title":"BIBM 2018 Copyright Page","authors":"","doi":"10.1109/bibm.2018.8621231","DOIUrl":"https://doi.org/10.1109/bibm.2018.8621231","url":null,"abstract":"","PeriodicalId":108667,"journal":{"name":"2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123707619","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}

{"title":"Analysis of multiscale entropy characteristics of heart rate variability in patients with permanent atrial fibrillation for predicting ischemic stroke risk","authors":"Ryo Matsuoka, K. Yoshino, E. Watanabe, K. Kiyono","doi":"10.1109/BIBM.2018.8621178","DOIUrl":"https://doi.org/10.1109/BIBM.2018.8621178","url":null,"abstract":"It has been reported that the complexity characteristics of heart rate variability (HRV) in patients with permanent atrial fibrillation (AFib) based on multiscale entropy (MSE) analysis are associated with ischemic stroke risk. However, the interpretation of HRV complexity is not clear and the mathematical and physical relationships between HRV and ischemic stroke have not been established. MSE is determined not only by the correlation characteristics but also by probability density function characteristics. The aim of this study was to clarify which characteristics were important for the association between MSE and ischemic stroke risk in patients with permanent AFib. We analyzed 24 hours of HRV data from 173 patients with permanent AFib. Results show that long-range correlations like 1/f fluctuations in a range greater than 90s were observed in HRV time series in patients with AFib, but that these values had no predictive power as an ischemic stroke risk factor. On the other hand, probability density functions of coarse-grained scales greater than 2s were significantly associated with ischemic stroke risk. These results suggest that probability density functions are a useful risk factor for improving ischemic stroke risk assessment. To investigate the probability density function characteristics more in detail, we analyzed the asymmetric non-Gaussian properties of the probability distribution of HRV data. Part of this study was published in the journal Entropy [1].","PeriodicalId":108667,"journal":{"name":"2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125370771","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}

{"title":"Classifying Osteosarcoma Using Meta-Analysis of Gene Expression","authors":"Olivia Alge, Jonathan Gryak, Yi-Yang Hua, K. Najarian","doi":"10.1109/BIBM.2018.8621119","DOIUrl":"https://doi.org/10.1109/BIBM.2018.8621119","url":null,"abstract":"Meta-analysis of gene expression provides the opportunity to compare gene expression across different platforms. In this paper, we use a meta-analysis of RNA-seq data collected by the SJTU team and publicly available microarray data to build a Random Forest classification model. The Random Forest model had average accuracy of 74.1% for cross-validation in the training set and achieved accuracy of 80.0% on the testing set.","PeriodicalId":108667,"journal":{"name":"2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125507896","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}

{"title":"Hypergraph regularized NMF by L2,1-norm for Clustering and Com-abnormal Expression Genes Selection","authors":"Na Yu, Ying-Lian Gao, Jin-Xing Liu, Juan Wang, J. Shang","doi":"10.1109/BIBM.2018.8621454","DOIUrl":"https://doi.org/10.1109/BIBM.2018.8621454","url":null,"abstract":"Non-negative matrix decomposition (NMF) has been widely used for sample clustering and feature selection in the field of bioinformatics. However, the existing methods based on NMF cannot effectively deal with the problem of intrinsic geometrical structure, noise, and outliers in gene expression data. In this paper, a novel method called Robust Hypergraph regularized Non-negative Matrix Factorization (RHNMF) is proposed to solve the above problem. Firstly, the hypergraph Laplacian regularization is introduced to consider the intrinsic geometrical structure of the high dimension data. Secondly, the L2,1-norm is applied in the error function to reduce effects of the noise and outliers, which may improve the robustness of the algorithm. Finally, we perform clustering and common abnormal expression genes (com-abnormal expression genes) selection on multi-view gene expression data to verify the rationality and validity of the RHNMF method. Extensive experimental results demonstrate that our proposed RHNMF method has better performance than other state-of-the-art methods.","PeriodicalId":108667,"journal":{"name":"2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126855692","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}

{"title":"Using multi-anchors to identify patients suffering from multimorbidities","authors":"Karl Øyvind Mikalsen, C. Soguero-Ruíz, I. Mora-Jiménez, Isabel Caballero Lopez Fando, R. Jenssen","doi":"10.1109/BIBM.2018.8621213","DOIUrl":"https://doi.org/10.1109/BIBM.2018.8621213","url":null,"abstract":"Chronic diseases, and in particular the co- occurrence of more than one chronic disease in an patient, which is known as multimorbidity, represent an increasing problem in modern society. For the individual patients the consequences are potentially serious, especially if not diagnosed and/or treated. In order to allow for an efficient allocation of health resources, and slow down the progression of these diseases, methods for predicting the health status of chronic patients have been developed. In this work, we propose a data-driven approach to identify the health status of chronically ill patients using data extracted from their electronic health records. For this purpose we take advantage of recent advances in machine learning for healthcare and use anchor learning that exploits vast amounts of unlabeled data. Moreover, in order to identify patients suffering from multimorbidities, we adapt the anchor method to a multi-anchor learning framework. The experiments show that using multi- anchor learning one can accurately identify patients who suffer from one or more chronic conditions. In fact, the performance is almost comparable to a completely supervised baseline.","PeriodicalId":108667,"journal":{"name":"2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115584900","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}

{"title":"A novel study for MDD detection through task-elicited facial cues","authors":"Jinlong Li, Zhenyu Liu, Zhijie Ding, Gangping Wang","doi":"10.1109/BIBM.2018.8621236","DOIUrl":"https://doi.org/10.1109/BIBM.2018.8621236","url":null,"abstract":"Depression is a common mental disorder worldwide. Individuals with major depressive disorder (MDD) are at increased risk for suicide. Current clinical practice for assessing this psychosomatic state are mainly based on self-report and expert evaluation, which risking a range of subjective biases. We investigate a number of task-elicited facial features from Chinese subjects for MDD detection. Moreover, we collect the data from Kinect, which make us achieve a good detection result with low time and space consumption. Experiments are performed on an age, gender and education level matched clinical dataset of 36 MDD patients and 36 healthy controls (HCs). We can get three points from the experimental results: 1) We have presented a simple and objective means for MDD detection, and the average classification accuracies (female: 71.5%, male: 66.7%) are all much higher than chance level. The best classification accuracies (female: 86.8%, male: 79.4%) are achieved during video watching task. 2) Neutral emotion stimulus is a better choice for data collection than positive and negative valences. 3) Eyebrows and mouth have more contributions than other parts of a face in neutral emotion valence. These findings suggest that detecting MDD from facial indicators is feasible, and we provide effective emotion stimulus and facial features.","PeriodicalId":108667,"journal":{"name":"2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"258 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116417579","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}

{"title":"A Selective Ensemble Learning Framework for ECG-Based Heartbeat Classification with Imbalanced Data","authors":"H. Ge, Keyi Sun, Liang Sun, Mingde Zhao, Chunguo Wu","doi":"10.1109/BIBM.2018.8621523","DOIUrl":"https://doi.org/10.1109/BIBM.2018.8621523","url":null,"abstract":"ECG-based heartbeat classification is often accompanied with difficult feature extraction and imbalanced sampling data. In order to alleviate the bias in performance caused by imbalanced data, a Selective Ensemble Learning Framework based on sample Distribution and classifier Diversity (SELFrame-DD) is proposed for ECG-based heartbeat classification. In SELFrame-DD, an improved SMOTE algorithm is proposed to generate training sets by using a sample-distribution based resampling strategy, and the selective ensemble depends on the diversity of classifiers and the prediction accuracy of classifiers for minority classes. Besides, a multimodal ECG feature extraction is employed based on wavelet packet decomposition and 1-D convolutional neural network. Experimental studies on MIT-BIH arrhythmia database show that the proposed algorithm can achieve a high classification accuracy for imbalanced multi-category classification.","PeriodicalId":108667,"journal":{"name":"2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114285180","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}

{"title":"Optimizing U-Net to Segment Left Ventricle from Magnetic Resonance Imaging","authors":"S. Charmchi, K. Punithakumar, P. Boulanger","doi":"10.1109/BIBM.2018.8621552","DOIUrl":"https://doi.org/10.1109/BIBM.2018.8621552","url":null,"abstract":"Left ventricle segmentation is an important medical imaging task to measure several diagnostic parameters related to the heart such as ejection fraction and stroke volume. Recently, convolutional neural networks (CNN) have shown great potential in achieving state-of-the-art segmentation results for such applications. However, most of the existing research is focusing on building complicated variations of the neural networks with modest changes to their performance. In this study, the popular U-Net architecture is optimized by analyzing its behaviour once fully trained from which one can simplify its architecture by fixing layers weights or eliminating some of them completely. For instance, by performing a Fourier analysis of the convolution at each layer, we were able to discover that some early layers can be approximated by simple uniform filters. Furthermore, in a separate experiment by removing the middle layers of the U-Net one can reduce the number of U-Net parameters from 31 million to 0.5 million weights without compromising its performance. The experimental evaluations show that the new optimized U-Net achieves 0.93 for the Dice score in comparison to manual ground truth.","PeriodicalId":108667,"journal":{"name":"2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114469497","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}

{"title":"An Integrative Framework for the Construction of Big Functional Networks","authors":"C. Giallombardo, Salvatore Morfea, Simona E. Rombo","doi":"10.1109/BIBM.2018.8621128","DOIUrl":"https://doi.org/10.1109/BIBM.2018.8621128","url":null,"abstract":"We present a methodology for biological data integration, aiming at building and analysing large functional networks which model complex genotype-phenotype associations. A functional network is a graph where nodes represent cellular components (e.g., genes, proteins, mRNA, etc.) and edges represent associations among such molecules. Different types of components may cohesist in the same network, and associations may be related to physical[biochemical interactions or functional/phenotipic relationships. Due to both the large amount of involved information and the computational complexity typical of the problems in this domain, the proposed framework is based on big data technologies (Spark and NoSQL databases). (0)","PeriodicalId":108667,"journal":{"name":"2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122079879","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}

{"title":"REPO-TRIAL: Common mechanism-based drug repurposing and endophenotyping","authors":"H. Schmidt","doi":"10.1109/BIBM.2018.8621156","DOIUrl":"https://doi.org/10.1109/BIBM.2018.8621156","url":null,"abstract":"Drug therapy and drug discovery are in a conceptual crisis. Hardly any new drug principles are discovered. Existing drugs have a catastrophic number needed to treat. Hardly any therapy targets a disease mechanism, because it is not known. Instead symptoms, biomarkers and risk factors are treated. Moreover we currently systemise medicine according to 19th and 20th century disease terms, which are mainly organ and symptom-based but not mechanistic. Network medicine utilizes common genetic origins, markers and co-morbidities to uncover mechanistic links between diseases. These links can be summarized in the diseasome, a comprehensive network of disease–disease relationships and clusters. The diseasome has been influential during the past decade, although most of its links are not followed up experimentally. We propose a new disease taxonomy based on mechanism and abolishing organ- and symptom-based disease definitions. Terms as hypertension, heart failure, arrhythmia will in future be considered mere disease phenotypes, most likely comprised of several endotypes and linked to several comorbities. Several such mechanistic clusters of disease phenotypes have been identified. One links to cyclic GMP and reactive oxygen species sources and targets. When examine the disease associations in a non-hypothesis based manner in order to identify possibly previously unrecognized clinical indications. Surprisingly, we find that sGC, the cardiovascular target of nitroglycerin, is closest linked to neurological disorders, an application that has so far not been explored clinically. Indeed, when investigating the neurological indication of this cluster with the highest unmet medical need, ischemic stroke, pre-clinically we find that sGC activity is virtually absent post-stroke. Conversely, a heme-free form of sGC, apo-sGC, was now the predominant isoform suggesting it may be a mechanism-based target in stroke. Indeed, this repurposing hypothesis could be validated experimentally in vivo as specific activators of apo-sGC were directly neuroprotective, reduced infarct size and increased survival. Thus, common mechanism clusters of the diseasome allow direct drug repurposing across previously unrelated disease phenotypes redefining them in a mechanism-based manner. Our example of repurposing apo-sGC activators for ischemic stroke should be urgently validated clinically as a possible first-in-class neuroprotective therapy and serves as a proof-of-concept for redefining disease, identifying new therapies. The REPO-TRIAL H2020 programme will develop an innovative in-silico based approach to improve the efficacy and precision of drug repurposing trials. We have chosen drug repurposing as it has the shortest time for clinical validation and translation. Validation of all putatively de novo discovered drug repositionings within the time-frame of this programme would be unrealistic. To improve efficacy and precision, and to adopt our computer simulation parameters and models","PeriodicalId":108667,"journal":{"name":"2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128691962","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}