Childhood Obesity Studies with Multicore Robust Data Mining Obesogenic Environment Variables Linear Regression Models

Childhood Obesity Studies with Multicore Robust Data Mining Obesogenic Environment Variables Linear Regression Models

Childhood Obesity Studies with Multicore Robust Data Mining Obesogenic Environment Variables Linear Regression Models

Kehoe, Bob, Contributor has reference to this Academic Journal, PHwiki organized this Journal Childhood Obesity Studies with Multicore Robust Data MiningProposal Review Meeting with CTSI Translating Research Into Practice Project Development Team, July 8, 2009, IUPUIGil Liu, Judy Qiu, Craig StewartContact Technology, UITSCommunity Grids Laboratory, PTIChildren’s Health ServiceIndiana University Obesogenic EnvironmentEnvironmental factors that increase caloric intake in addition to decrease energy expenditure “ so manifold in addition to so basic as to be inseparable from the way we live.” Margaret Talbot (New America Foundation)“The current U.S. environment is characterized by an essentially unlimited supply of convenient, inexpensive, palatable, energy-dense foods coupled with a lifestyle requiring negligible amounts of physical activity as long as subsistence.” Hill & Peters 2001“Genes load the gun, in addition to environment pulls the trigger.” G Bray 1998

Sumter Beauty College SC

This Particular University is Related to this Particular Journal

of Visits Per patient Percent 1 only 44% 2 or more 46% 3 or more 22% 4 or more 11% 5 or more 6%Distribution of Visits by Year in addition to FrequencyYear of visits2004 43005 2005 452712006 453002007 54707 Zones of Analysis Centered on Subject’s Residence

units/acrevery low density 0-2low density 2-5medium density 5-15high density > 15commercial lightcommercial officecommercial heavyindustrial lightIndustrial heavyspecial useparksroadswaterinterstatesGeneralized L in addition to Use Categories012Milesvacant / agriculturalThe Environment GREENNESS Normalized Difference Vegetation Index (NDVI) Healthy green biomassVariables of the Built Environment Selected as long as Study:VariablesDependent2-year change in BMI z-Score (t2-t1)CovariatesAge, race/ethnicity, sex Baseline z-BMI (linear, quadratic, cubic) Health insurance statusCensus tract median family income (log)Index year

Linear Regression Models of 2-year change in z-BMIPotential Pathways in addition to MechanismsPlaces that promote outside play in addition to physical activity“Territorial personalization”Improved mental health, self-esteem, reduced stressCollaboration of SALSA ProjectIndiana University ITSALSA TeamGeoffrey Fox Xiaohong QiuScott BeasonSeung-Hee BaeJaliya Ekanayake Jong Youl ChoiYang RuanMicrosoft ResearchIndustry Technology Collaboration DryadRoger BargaCCRGeorge ChrysanthakopoulosDSSHenrik Frystyk NielsenApplication CollaboratorsBioin as long as matics, CGB Haiku Tang, Mina Rho, Qufeng DongIU Medical School Gilbert LiuIUPUI Polis Center (GIS) Neil DevadasanChemin as long as matics Rajarshi Guha, David Wild PTI/UITS RTCraig Stewart William BernnetScott Mcaulay

DataDeveloping in addition to applying parallel in addition to distributed Cyberinfrastructure to support large scale data analysis. Childhood Obesity Studies (314,932 patient records/188 dimensions) Indiana census 2000 (65535 GIS records / 54 dimensions) Biology gene sequence alignments (640 million / 300 to 400 base pair) Particle physics LHC (1 terabytes data that placed in IU Data Capacitor)Components of Data Intensive Computing SystemComponents of Data Intensive Computing SystemHardwareThe exponentially growing volumes of data requires robust high per as long as mance tools. Parallelization frameworks MPI as long as High per as long as mance clusters of multicore systems MapReduce as long as Cloud/Grid systems (Hadoop , Dryad) Data mining algorithms in addition to tools Deterministic Annealing Clustering (VDAC) Pairwise Clustering Multi Dimensional Scaling (Dimension Reduction) Visualization (Plotviz)Components of Data Intensive Computing SystemSoftware

Data Intensive (Science) Applications Heath Biology Chemistry Particle Physics LHC GISComponents of Data Intensive Computing SystemApplicationDeterministic Annealing Clustering of Indiana Census DataDecrease temperature (distance scale) to discover more clustersDistance Scale Temperature0.5Red is coarse resolution with 10 clustersBlue is finer resolution with 30 clustersClusters find cities in IndianaDistance Scale is Temperature Various Sequence Clustering Results184500 Points : Clustal MSAMap distances to 4D Sphere be as long as e MDS3000 Points : Clustal MSA Kimura2 Distance

Initial Obesity Patient Data Analysis19Refinement of 3 of clusters to left into 54000 records 8 ClustersJune 11 2009June 11 2009Parallel Overhead

Pairwise Sequence Distance CalculationPer as long as m all possible pairwise sequence alignment given a set of genomic sequences.Alignments per as long as med using Smith-Waterman (local) sequence alignment algorithm.Currently we are able to per as long as m ~640 million alignments (300 to 400 base pairs) in ~4 hours using tempest cluster.Represents one of the largest datasets we have analyzed.MDS of 635 Census Blocks with 97 Environmental PropertiesShows expected Correlation with Principal Component – color varies from greenish to reddish as projection of leading eigenvector changes valueTen color bins usedCanonical CorrelationChoose vectors a in addition to b such that the r in addition to om variables U = aT.X in addition to V = bT.Y maximize the correlation = cor(aT.X, bT.Y).X Environmental DataY Patient DataUse R to calculate = 0.76

Kehoe, Bob DRIVE! Magazine Contributor

Projection of First Canonical Coefficient between Environment in addition to Patient Data onto Environmental MDSKeep smallest 30% (green-blue) in addition to top 30% (red-orchid) in numerical value Remove small values < 5% mean in absolute valueMDS in addition to Canonical CorrelationReferencesSee K. Rose, "Deterministic Annealing as long as Clustering, Compression, Classification, Regression, in addition to Related Optimization Problems," Proceedings of the IEEE, vol. 80, pp. 2210-2239, November 1998T Hofmann, JM Buhmann Pairwise data clustering by deterministic annealing, IEEE Transactions on Pattern Analysis in addition to Machine Intelligence 19, pp1-13 1997Hansjörg Klock in addition to Joachim M. Buhmann Data visualization by multidimensional scaling: a deterministic annealing approach Pattern Recognition Volume 33, Issue 4, April 2000, Pages 651-669Granat, R. A., Regularized Deterministic Annealing EM as long as Hidden Markov Models, Ph.D. Thesis, University of Cali as long as nia, Los Angeles, 2004. We use as long as Earthquake predictionGeoffrey Fox, Seung-Hee Bae, Jaliya Ekanayake, Xiaohong Qiu, in addition to Huapeng Yuan, Parallel Data Mining from Multicore to Cloudy Grids, Proceedings of HPC 2008 High Per as long as mance Computing in addition to Grids Workshop, Cetraro Italy, July 3 2008Project website:

Kehoe, Bob Contributor

Kehoe, Bob is from United States and they belong to DRIVE! Magazine and they are from  Orange, United States got related to this Particular Journal. and Kehoe, Bob deal with the subjects like Regional Interest

Journal Ratings by Sumter Beauty College

This Particular Journal got reviewed and rated by Sumter Beauty College and short form of this particular Institution is SC and gave this Journal an Excellent Rating.