Day 2 :
Keynote Forum
Ramalingam Shanmugam,
Texas State University, USA
Keynote: Correlation and distance concepts together suggest impaired driving due to alcohol than marijuana is menace to road safety
Time : 09:45-10:35
Biography:
Ram Shanmugam is the Editor-in-Chief for the journals: Epidemiology & Community Medicine, Advances in Life Sciences and Health, and Global Journal of Research and Review. He is the Associate Editor of the International Journal of Research in Medical Sciences. He is the Book-Review Editor of the Journal of Statistical Computation and Simulation. He directed Statistics Consulting Center in the Mississippi State University. In 2015, he has published a textbook with the title “Statistics for Engineers and Scientists”. He served the Argonne National Lab., University of Colorado, University of South Alabama and the Indian Statistical Institute. He has published 125 research articles and is a Fellow of the International Statistical Institute. Currently, he is a Professor in the School of Health Administration, Texas State University. He is a recipient of several research awards from the Texas State University
Abstract:
Sometimes, repeated experience with accidents makes people to believe and conclude that impaired driving due to alcohol or marijuana is menace to road safety. However, policy makers on road safety require data evidence authenticated interpretations. What statistical methodologies now exist to do so to help policy makers? For this purpose, this exploratory and tutorial article is written with data analyses and interpretations of actual number of fatal accidents caused by impaired driving during 2013-2015 in USA (as reported in Arnold and Teft 2016). This research work first convinces, using regression analysis, that the driver’s age is not a significant predictor of fatal accidents. Then, in a novel manner, it mixes correlation and Mahalanobis distance concepts to create an approach to check whether impaired driving due to alcohol or marijuana is a serious menace to the road safety. In conclusion, this research work finds that just by eliminating impaired drivers due to alcohol (but not marijuana) could ensure road safety 16.02 times closer to an ideal situation of no fatal road accident
Keynote Forum
Mikhail Moshkov
King Abdullah University of Science and Technology, Saudi Arabia
Keynote: Extensions of dynamic programming for machine learning and knowledge representation
Time : 10:55-11:40
Biography:
Mikhail Moshkov is Professor in the CEMSE Division at King Abdullah University of Science and Technology, Saudi Arabia since October 1, 2008. He earned Master’s degree from Nizhni Novgorod State University, received his Doctorate from Saratov State University, and Habilitation from Moscow State University. From 1977 to 2004, he was with Nizhni Novgorod State University. Since 2003, he worked in Poland in the Institute of Computer Science, University of Silesia, and since 2006 also in the Katowice Institute of Information Technologies. His main areas of research are complexity of algorithms, combinatorial optimization, and machine learning. He is author or co-author of five research monographs published by Springer
Abstract:
We discuss so called multi-pruning which allows us to construct classifiers (decision trees) that outperform often classifiers constructed by CART. This approach is based on the construction of the set of Pareto optimal points for bi-criteria optimization problem relative to the size of decision trees and the number of misclassifications. The second topic is connected with multi-stage optimization of decision rules relative to the coverage and length. Based on this optimization procedure, we can simulate the work of greedy algorithm for the set cover problem. As a result, for many datasets from UCI ML Repository, we can construct small systems of enough accurate decision rules that cover the most part of objects (rows). The end of the presentation is devoted to the introduction to KAUST
- Modern data analysis|Biometric security|Regression Analysis|Biostatistics applications|Bayesian statistics|Clinical Biostatistics|Adaptive biometrics systems
Location: Houston, USA
Chair
Yedidi Narasimha Murty
Electronic Arts, USA
Co-Chair
Mikhail Moshkov
King Abdullah University of Science and Technology, Saudi Arabia
Session Introduction
Leigh Anne H Clevenger
Pace University, USA
Title: Review of biometric classification of heart sound for continual user authentication and clinical applications
Time : 11:50-12:20
Biography:
Leigh Anne H Clevenger is a candidate of Doctor in Professional Studies in Computing at Pace University. As a software engineer at IBM in Poughkeepsie, NY, she has developed solutions for advanced technology microprocessor design for six technology generations. She has nine submitted patents in the area of wearables and healthcare. She was an invited speaker at the 2015 Pace University Cybersecurity Workshop
Abstract:
Continual authentication using passive monitoring of sensor data is not currently available on most mobile devices. This monitoring can maintain confidence that the device owner is the current user without inconveniencing them by requiring frequent re-authentication, for example with password, swipe, or fingerprint. Biometrics used for passive monitoring do not currently include heart sound, which is an interesting choice because it is constantly available, hard to obtain from another person, and has been shown to be reasonably unique between individuals. Clinical cardiology applications currently do not take advantage of the algorithms of heart sound authentication, for example, to indicate a change in the patient’s heart sound on an in-home wearable mobile device app. This research explores the biometric of heart sound for use in passive and continual screening for clinical applications, and for user authentication. Using the heart sound biometric for a cardiac patient allows passive monitoring of sensor data, screening changes in heart sound. Changes from baseline data trigger an alert to the user and caregiver. For user authentication, passive monitoring maintains confidence that the device owner is the current user without inconveniencing users by asking them to re-authenticate to access high security applications. Prior heart sound research is extended for potentially greater user authentication accuracy in the areas of time windows, number of heartbeats, feature vectors, classifiers, sample selection, and noise mitigation. Application and adaptation of user authentication methodologies from speech processing are applied. The methodology can be extended to work with different public and private heartsound datasets
Shaikh Mohammad Bokhtiar
SAARC Agriculture Centre, Bangladesh
Title: Reliability and policy framing for fisheries statistics in saarc region
Time : 12:20-12:50
Biography:
S M Bokhtiar was born in Chapai Nawabganj district, Bangladesh on 1 January, 1963. He graduated and achieved B.Sc. Ag (Hons.) from Bangladesh Agricultural
University (BAU), Mymensingh in 1985. MS in soil science from Bangabandhu Sheikh Mojibur Rahman Agricultural University (BSMRAU), Bangladesh and Ph.D
degree from the United Graduate School of Agricultural Science, Ehime University, Japan in 1999 and 2006, respectively. Dr. Bokhtiar worked as post doctoral
research fellow at Guangxi Academy of Agricultural Sciences, Guangxi, China for two years and studied on silicon nutrition of sugarcane crop. He started his carrier
as a Scientific Officer in Farming Systems Research and Development Project (FSR & D) at Bangladesh Sugarcane Research Institute (BSRI), Bangladesh in
1989. During his service period at BSRI, Dr. Bokhtiar was promoted as a Senior Scientific Officer and also performed as a heads of division of Soils & Nutrition
Division, BSRI till December 2010. Dr. Bokhtiar was appointed as a Principal Scientific Officer at Soils Unit of Natural Resources Management Division of
Bangladesh Agricultural Research Council (BARC), Bangladesh in 10 January 2011 and assigned for programme planning, execution, evaluation and monitoring of
soils programme of National Agricultural Research Systems (NARS) in Bangladesh. Currently Dr. Bokhtiar serving as a Director, SAARC Agriculture Center (SAC),
Dhaka, Bangladesh and involved in policy planning, formulation and implementation of the activities in the SAARC member states assigned by SAARC Secretariat
Katmandu Nepal. Dr. Bokhtiar has 60 research papers in his credit with total citations of 253 and author of two books. Dr. Bokhtiar attended several international
seminars and training programme in home and abroad. Dr. Bokhtiar visited several countries like Japan, Thailand, China, Egypt, Philippine, Malaysia, Mongolia,
South Korea, Pakistan, New Zealand and India. Dr. Bokhtiar is actively associated with the International Association of Professionals in Sugar and Integrated
Technologist (IAPSIT) based in Nanning, China since the very beginning of its formation in 2004. Dr. Bokhtiar also served as a Member-Secretary of Exchange and Cooperation Consortium for Agricultural Science and Technology, China- South Asia (ECCAST-CSA) Bangladesh part.
Abstract:
World fisheries production have remarkably increased since 1950 and with present annual 167.2 million tonnes fish production (FAO, 2016), fisheries and aquaculture became the potential contributors to food and nutrition security and livelihoods at global level. Almost 90% of aquaculture production takes place in Asia, most of it in the tropical and subtropical countries. The two South Asian Association for Regional Cooperation (SAARC) countries, India and Bangladesh with the annual production of over 10.0 and 3.55 million tonnes in 2016, respectively rank the 2nd and 5th largest fish producers in the world. The sector employs 56.6 million people globally of which India and Bangladesh alone share 32 million people. In South Asian region, at present hardly one third of the existing freshwater ponds and water bodies are engaged in aquaculture. Most of the rural people in the region depend on their backyard ponds and seasonal ponds for their house hold fish requirement throughout the year. These fisheries catches contribute substantially to the national fish production data. However, these production data are never included in the respective nation’s fish statistics data base. Therefore, there is underestimation of fish production data for any particular country in SAARC region. At present, the Food and Agriculture Organization of the United Nations (FAO) maintains global fisheries statistics by collecting data from the member countries. Past experience shows FAO sometimes encountered with incorrect data. Therefore, fisheries data may be scrutinized by the various regional bodies before sending them to world data pool. In this regards, SAARC can play the leading role for regional data pulling and scrutinisation. Also, review of existing methodologies for fish production estimation from diverse water bodies need serious attention. Based on the outcome, necessary policy may be framed at SAARC regional level for fisheries data collection and accurate reporting. Without reliable statistics, effective fisheries management and policy-making are impossible in the region, the major contributor to global fisheries production
Tatsuya Takagi
Osaka University, Japan
Title: Predicting hydrolyzability using logistic regression analyses and regularization techniques
Time : 13:40-14:10
Biography:
Tatsuya Takagi has completed his PhD from Osaka University. At that time, he had been an Assistant Professor of School of Pharmaceutical Sciences, Osaka University for 5 years. Then, since 1993, he had worked for the Genome Information Research Center, Osaka University as an Associate Professor until he became a Professor of Graduate School of Pharmaceutical Sciences, Osaka University in 1998. He has published more than 100 papers in reputed journals and serving as Chairman of Division of Structure-Activity Relationship of the Pharmaceutical Society of Japan.
Abstract:
It is significant to estimate the environmental fates of chemical substances which are emitted from factories or as residential wastes. Especially, since hydrolysis plays a main role with regard to chemical substance degradation in the environment, hydrolyzability of such chemicals have to be revealed. However, experimentally obtaining the information is time-consuming. Thus, we tried to predict the hydrolyzability of esters and related compounds using logistic regressions and regularization methods. The hydrolyzability data of 143 chemicals, which were extracted from literatures, were used for these analyses. These chemicals were classified into two categories, ‘stable’ and ‘hydrolizable’, according to their half-life periods. They were also classified into four groups, all chemicals (143), esters (73), amides, and others. In this study, the former two groups were analysed. 88 chemical descriptors were prepared for predicting the hydrolyzability. All the datasets were divided into training (3/4) and test (1/4) sets. Lasso was used as a regularization method. We built the model equation by two techniques using only training data sets. As the results of the analyses, training data were perfectly predicted in the case of esters, and sufficient results were obtained in the case of all chemicals. Even in the case of test data sets, satisfactory results were obtained.
Zeleke Worku
Tshwane University of Technology Business School, South Africa
Title: Predictors of adverse outcomes of pregnancy in south african women
Time : 14:10-14:40
Biography:
Professor Zeleke Worku is a South African academic working at the Business School of Tshwane University of Technology (TUT) in Pretoria, South Africa as an associate professor of statistics and coordinator of the MBA programme of study at TUT Business School. He holds a Ph.D. in statistics (University of the Orange Free State in Bloemfontein, South Africa) and a second Ph.D.in sociology (Aalborg University, Denmark). Professor Worku’s key research interests are in monitoring and evaluation, statistical data mining, biostatistics, epidemiology, public health, sociology, demography, econometrics and business sciences. Before he joined TUT Business School in 2010, Professor Worku has served the University of Natal in Durban, South Africa (1998 to 1999), Vista University in Pretoria, South Africa (2000), the University of Pretoria, Pretoria, South Africa (2001 to 2007), and the University of South Africa in Pretoria, South Africa (2008 to 2009). Professor Worku lives and works in Pretoria, South Africa with his wife and two children
Abstract:
A review of the relevant literature shows that teenage pregnancy and adverse outcomes of pregnancy constitute a major public health problem in South African women of the childbearing age of 15 to 49 years. A longitudinal study was conducted in Tshwane, South Africa in order to identify factors that affect utilization of modern contraceptives and adverse pregnancy outcomes in women of the childbearing age of 15 to 49 years. Data analysis was conducted by using statistical methods such as binary logistic regression analysis, survival analysis, multilevel analysis and Bayesian analysis. The study showed that the percentage of women who regularly used modern family planning methods such as condoms, pills, injections, intra-uterine devices and sterilization was 41.74%. The average ages of women at first sex and pregnancy were 18.72 and 19.36 years respectively. Adverse outcomes of pregnancy occurred in 12.19% of women. Based on Odds Ratios (OR) estimated from binary logistic regression analysis, utilization of contraceptives was significantly influenced by easy access to family planning services, level of support from sexual partner, and young age at first pregnancy. Based on hazard ratios (HR) estimated from the Cox Proportional Hazards Model, the occurrence of adverse outcomes of pregnancy was significantly influenced by easy access to family planning services, unwanted pregnancy, and young age at first pregnancy. Women who experienced adverse outcomes of pregnancy were characterized by poor utilization of reproductive health and modern family planning services. Based on results estimated from multilevel analysis, there was a significant difference among the 20 health service delivery wards and 11 health service facilities in which reproductive health services were delivered to women with regards to the quality of service delivery
Tao Liu
Brown University School of Public Health, USA
Title: Date driven method for optimal allocation of gold standard testing under constrained availability
Time : 14:40-15:10
Biography:
Tao Liu has completed his PhD from the University of Pennsylvania. He is an Assistant Professor at Brown University, Associate Director of Data and Statistics Core of the Alcohol Research Center on HIV (ARCH) and a faculty member of the Center for Statistical Sciences (CSS) and Center for AIDS Research (CFAR). His research expertise includes “Design of clinical trials, clinical decision making, analysis of incomplete data, sensitivity analysis, and statistical causal inference”. His collaborative research interest focuses on the area of HIV/AIDS and related diseases
Abstract:
The World Health Organization (WHO) guidelines for monitoring the effectiveness of human immunodeficiency virus (HIV) treatment in resource-limited settings are mostly based on clinical and immunological markers (e.g., CD4 cell counts). Recent research however indicates that the guidelines are inadequate and can result in high error rates. Viral load (VL) is considered the “gold standard,” yet its widespread use is limited by cost and infrastructure. In this talk, a two-step diagnostic algorithm is presented that uses information from routinely collected clinical and immunological markers to guide a selective and targeted use of VL testing for diagnosing HIV treatment failure, under the assumption that VL testing is available only at a certain portion of patient visits. The proposed algorithm identifies the patient subpopulation, such that the use of limited VL testing on them minimizes a predefined risk (e.g., misdiagnosis error rate). Diagnostic properties of our proposed algorithm are demonstrated by simulations. The method is illustrated using data from an HIV clinic in Rhode Island, and results show considerable promise for improving the effectiveness of HIV treatment monitoring in resource limited settings
Basiru Yusuf
Jigawa State Polytechnic, Nigeria
Title: Regression analysis of large data set “Partial least squares regression with application to analysis of chemometric dataâ€
Time : 17:20-17:50
Biography:
Basiru Yusuf has completed his MSc in Statistics from University of Leeds UK and currently pursuing his PhD. He is a Principal Instructor I at Statistics Department in Jigawa State Polytechnic Dutse. He has conducted many researches, published conference papers and engaged in teaching, guiding to research project and industrial visits.
Abstract:
Regression analysis of large data sets (multivariate data) such as chemometric and microarray data which has more variables than the number of observations (n<
- YRF
Location: Houston,USA
Session Introduction
Adelino Martins
Eduardo Mondlane University, Mozambique
Title: A new model for multivariate current status data
Time : 15:10-15:35
Biography:
Adelino Martins has completed his Master degree from Hasselt University and pursuing PhD at Hasselt University. He was a Lecturer at Eduardo Mondlane University in Maputo, Mozambique.
Abstract:
Individual heterogeneity in the acquisition of infectious diseases is recognized as a key concept, which allows improved estimation of important epidemiological parameters. Frailty models allow to represent such heterogeneity. Coull (2006), introduced a computational tractable multivariate random effects model for clustered binary data. The objective of this report was to apply and modify the proposed model, and compare to the shared and correlated gamma frailty models in the context of the analysis of multivariate current status data. The models were applied to the bivariate current status data on Varicella-Zoster Virus and Parvovirus B19 using different baseline hazard functions for the force of infection. The findings revealed that the proposed model which is called in this report as new correlated gamma frailty model is closely related to existing frailty models. The main difference is the way the multivariate gamma is introduced in the model, and the indirect way to specify the baseline hazard function. In terms of construction, a frailty model is typically formulated based on the specification of the proportional hazard function, whereas the new correlated gamma frailty model is built using a classical generalized linear mixed model for clustered binary data. Furthermore, in the new model the variances of the frailties are assumed to be identical, whereas in case of the frailty model, the variances can be different or identical and the correlation is constrained by the ratio of the variances.
Abolade Olawale
Osun State Polytechnic, Nigeria
Title: Prevalence and survival determinants of cancer in Nigeria
Time : 15:35-16:00
Biography:
Mrs Olawale holds M.Sc degree in Statistics.She currently lectures at Osun State Polytechnic, Iree Nigeria. Mrs Olawale has presented several papers both locally and internationally. Presently, she is a PhD student in the department of Statistics,University of Ilorin Nigeria.Her research area is Bio-statistics with interest in Survival Analysis
Abstract:
Data collected in South West Nigeria, which covers about 25% of Nigeria land mass typically shows wide spread of this disease across age-groups. However, the cox survival analysis of the data has shown that age is the main risk factor within each type of cancer for death while breast cancer constitute more than one quarter of its prevalence. Other noticeable cancer types include liver, rectum, blood, ovary, skin, prostrate and pancreas. The uncommon cancers include Epigastric, nasopharyngeal, gall bladder, bone and brain. This indicates that while previously common types of cancer still exist, cancers such as nasopharyngeal and lymph are now showing presence in that part of the world
- Workshop
Location: Houston,USA
Session Introduction
Upendra Kumar Devisetty
University of Arizona, USA
Title: Bringing your favorite bioinformatics analysis tools to cyverse using docker
Time : 16:20-17:20
Biography:
Upendra Kumar Devisetty earned his PhD from University of Nottingham, UK and completed Post-doctoral studies at the University of California Davis and Oregon State University. He is currently working as Science Informatician at CyVerse, a life sciences cyberinfrastructure funded by the National Science Foundation (NSF). He has published more than 10 papers in peer-reviewed journals and has been invited to speak at several international conferences
Abstract:
CyVerse (formerly iPlant Collaborative) is a life sciences cyberinfrastructure funded by the National Science Foundation (NSF). The infrastructure’s purpose is to scale science, domain expertise, and knowledge by providing a variety of computational tools, services, and platforms for storing, sharing, and analyzing large and diverse biological datasets. The Discovery Environment (DE) in CyVerse specifically provides a modern web interface for running powerful computing, data, and analysis applications. By providing a consistent user interface (UI) for accessing applications and computing resources needed for specialized scientific analyses, the DE facilitates data exploration and scientific discovery. DE merges the “science gateway” functionality and the bioinformatics “workbench” with high-performance data management to allow seamless access to reusable computational workflows that can run at very large scales. It is common in bioinformatics to build new analysis methods utilizing multiple programs, libraries, and modules. However, each analysis that uses these tools requires specific versions of the operating system and underlying software. Docker is a container virtualization technology that wraps a bioinformatics tool (e.g BWA) together with all its software dependencies so it can run in a reproducible manner irrespective of enviroment. This workshop will teach users how to install Docker, write a Docker file for their bioinformatic tool of interest, build the Docker image containing the tool, test the built Docker image, submit a tool request, build the new app UI in the DE and finally test their web app and share it with their collaborators or make it public so that other users can use it