Computational Learning Approaches to Data Analytics in Biomedical Applications

Abstract:          

The advancement of translational medical research to achieve personalized medicine has increased demand for computational intelligence tools that can perform robust analysis, rapid interpretation and learn solutions from the increasing amount of data generated by biomedical applications.  Machine learning involves both supervised learning (classification) and unsupervised learning or exploratory data analysis (clustering).

There are many rich theories and applications of computational learning approaches to data analytics arising from a wide variety of communities, ranging from engineering and computer sciences (computational intelligence, data mining, information retrieval, machine learning, pattern recognition), life and medical sciences (biology, clinic, genetics, microbiology, paleontology, pathology, psychiatry, phylogeny), and earth sciences to social sciences (anthropology, psychology, sociology) and economics (business, marketing). Such diversity causes confusion because of differing terminologies, goals and lack of good communication between communities. This tutorial will help the audience understand this field as it relates to biomedical applications and translational medicine.

We will begin with an introduction and discussion of the major problems relating to data analytics in biomedical applications. We will provide a review of important, influential, and state-of-the-art learning algorithms in literature for biomedical applications with special emphasis on clustering algorithms. We will review cluster validity indices and discuss how to select the appropriate one for one’s data. The tutorial will include hands-on training on use of intuitive software tools. We encourage participants to come prepared with a brief synopsis of what type of biomedical data they are interested in. They can bring their own data or we will have sample datasets that relate specifically to phenotype medical datasets and genomic data.

Outline:

  • Overview of data analytics in biomedical applications and current challenges
  • Supervised learning algorithms and applications
  • Clustering algorithms
  • Cluster validation indices
  • Computational tools for biomedical data analysis
  • Hands-on training on use of Python libraries, WEKA, SAP-HANA, R/Bioconductor, JMP

List of Speakers 

Donald Wunsch II, Missouri University of Science and Technology (Missouri S&T), dwunsch@mst.edu
Title of the presentation: Computational learning approaches in biomedical applications: overview, theories & challenges
Bio:  Dr. Wunsch received his Ph.D. in Electrical Engineering from the University of Washington, Seattle. He is the Mary K. Finley Missouri Distinguished Professor, Missouri University of Science and Technology (S&T). His research interests include clustering, adaptive resonance & reinforcement learning architectures (hardware and applications), bioinformatics.

Tayo Obafemi-Ajayi, Missouri S&T/Missouri State University, towd2@mst.edu
Title of the presentation: Cluster validation indices and software tools for biomedical data analysis
Bio: Dr. Obafemi-Ajayi is an Assistant Professor of Electrical Engineering at Missouri State University in the Cooperative Engineering Program, a joint program with S&T. She recently completed a post-doctoral fellowship with the Applied Computational Intelligence Lab at S&T under the supervision of Dr. Wunsch, working on clustering and genomic data analysis related to Autism. She obtained her PhD in Computer Science from Illinois Institute of Technology. Her research interests are machine learning, bioinformatics, and data mining.

Gayla Olbricht, Missouri S&T, olbrichtg@mst.edu
Title of the presentation: Statistical analysis and software tools for biomedical data analysis
Bio: Dr. Olbricht is an Assistant Professor in the Department of Mathematics and Statistics at Missouri S&T. She earned her Ph.D. in Statistics from Purdue University. Her research interests include Markov models, regression analysis, statistical genomics, and bioinformatics.

B. Khalid Al-Jabery, Missouri S&T, kkatnb@mst.edu
Title of the presentation: Hands on training on use of Python libraries, WEKA, SAP-HANA, R & JMP
Bio: Khalid Aljabery is currently a PhD candidate and research assistant at the Missouri S&T. He obtained his BS and M.Sc. in Computer Engineering at the University of Basrah in Iraq in 2005 and 2009 respectively. He has more than 6 years of experience as an IT engineer. He worked for ExxonMobil, South Oil Company-Iraq, and International Organization of Migration (IOM). He has gained hands-on experience on applying a variety of computational tools to machine learning projects on power optimization, smart grid, and medical datasets.