2nd Seminar by CAST—Centre for Applied Statistics and Data Analytics
Dates: August 17th, 2016 (1h15 p.m.- 4 p.m.)
Venue: University of Tampere, Finland
Local Organisation: Ansa Lilja: Ansa.Lilja@uta.fi
The second Seminar by the Centre for Applied Statistics and Data Analytics (CAST) will be held at the School of Health Sciences, University of Tampere, Finland, at August 17th, 2016. It will gather researchers, other faculty and students interested in applied statistics and data analytics working at the University of Tampere and other Universities, Research Institutes and Companies.
The main aims of the seminar events by CAST are: (i) to bring awareness of the importance of statistics and data analysis in research; (ii) to create a forum of discussion where researchers present their work and research questions followed by discussion and feedback from the audience; (iii) strengthen the links between schools and research groups that might lead to future collaborations in terms of research articles and funding applications.
The visiting speakers of the seminar are Professor Dulal Bhaumik and Research Assistant Professor Runa Bhaumik from University of Illinois at Chicago. Professor Dulal Bhaumik has worked both on applied and theoretical statistics. The talk given by him will be based on his applied work, mainly neuroimaging and neuroconnectivity. He will use fMRI and DTI data to show how brains malfunction for neurological diseases which is very useful to treat patients. Research Assistant Professor Runa Bhaumik in turn is a Computer Scientist who works in the same area as Dulal. Her interest is in image processing. She uses machine learning techniques (such as Support Vector Machine, Random Forest etc.) for classification of subjects with various neurological conditions.
Please spread the information to whom it may concern and register for the event by Monday 15th, at 1:00 p.m.
Wednesday, August 17th, 2016 (A207 lecture hall, Arvo building in Kauppi)
13h15 – 13h30: Opening
13h30 – 14h45: Dulal K. Bhaumik : Neuro-connectivity analysis for autistic subject utilizing fMRI data
14h45 - 15h15: Coffee break
15h15 – 16h00: Runa Bhaumik : Predictive Modeling Using Big Data in Health Informatics
Neuro-connectivity analysis for autistic subjects utilizing fMRI data
Dulal K. Bhaumik, PhD
Director, Biostatistical Research Center
Professor of Biostatistics, Psychiatry and Bioengineering
University of Illinois at Chicago, USA
Several recent studies provide evidence that neural network is over-connected in autism, but not always under-connected as previously thought. These apparent discrepancies can be attributed to some drawbacks in study designs such as small sample size, failure to account for subject variations etc. We explore the largest autism database, the Autism Brain Imaging Data Exchange (ABIDE) and analyze 361 subjects from 8 medical centers, thus avoid the ‘underpower’ issue. In addition, we are able to account for subject variations by fitting mixed-effects models via EM-like algorithm. Principal factor approximation is used to control the false discovery proportion (FDP) for correlated hypotheses.
Predictive Modeling Using Big Data in Health Informatics
Runa Bhaumik, University of Illinois at Chicago, USA
Abnormal resting-state functional connectivity of distributed brain networks may aid in understanding mechanisms involved in major depressive disorder (MDD). To date, few studies have used resting state functional magnetic resonance imaging (rs-fMRI) to attempt to discriminate individuals with MDD from individuals without MDD, and to our knowledge no investigations have examined a remitted (r) population. In this study, we examined the efficiency of L1-norm support vector machine (SVM) classifier to successfully discriminate unmedicated rMDD individuals from healthy controls (HCs) in a narrow early adult age range. We empirically evaluated four feature selection methods including multivariate least absolute shrinkage and selection operator (LASSO) and Elastic Net feature selection algorithms. Our results showed that SVM classification with Elastic Net feature selection achieved the highest classification accuracy of 76.1% (sensitivity of 81.5% and specificity of 68.9%) across subjects from a dataset consisting of 38 (r) MDD individuals and 29 healthy controls. The highest discriminating functional connections were between the left amygdala, left posterior cingulate cortex, bilateral dorso-lateral prefrontal cortex, and right ventral striatum. These appear to be key nodes in the etiopathophysiology of MDD, within and between default mode, salience and cognitive control networks. This technique demonstrates early promise for using rs-fMRI connectivity as a putative neurobiological marker capable of distinguishing between individuals with and without rMDD. These methods may be extended to periods of risk, prior to illness onset, thereby allowing for earlier diagnosis, prevention, and intervention.