Director & Full Member, Biostatistics Core, Research Institute
Professor of Bioinformatics and Biostatistics, Institute for Academic Medicine
Weill Cornell Medical College
Dr. Peterson’s dissertation research at the University of Texas-Human Genetics Center (Director, William J. Schull) was funded by the U.S. Nuclear Regulatory Commission and focused on statistical models of error in radiation dosimetry and cancer diagnosis among Hiroshima and Nagasaki atomic bomb survivors, and projection of lifetime mortality risks of radiation-induced cancer among US nuclear workers.
Dr. Peterson’s previous appointments included Shuttle Mission Radiation Specialist at the Lyndon B. Johnson Space Center of the National Aeronautics and Space Administration as well as Associate Professor of Medicine (Primary) and Associate Professor of Molecular and Human Genetics (secondary) at Baylor College of Medicine. Dr. Peterson also holds adjunct appointments in the Division of Biostatistics, School of Public Health, University of Texas, and Section of Cardiology, Department of Medicine, Baylor College of Medicine.
As Director of the Center for Biostatistics at the Research Institute, he collaborates in a wide range of multidisciplinary-translational research collaborations involving development of biologically-motivated carcinogenic risk mitigation models for space radiation exposure (NASA Space Radiation Program), computational pathway abrogation for identification of genes predictive of breast cancer survival (Lawrence Berkeley National Lab), development of a saliva-based nano-biochip for diagnosis of acute myocardial infarction (Rice University, BCM, Houston-VAMC), data mining outcomes of spinal cord injury (HM Neurosurgery), and design and analyses of molecular and clinical studies metabolic syndrome disorders and diabetes (HMRI Diabetes and Metabolic Disease).
Dr. Peterson has been a member of several NCI EDRN and SPORE review panels, molecular-genetics reviews for the Dept. of Defense research program, and behavioral genetics reviews for NIH-BGES. In the area of computational intelligence (neural networks), Dr. Peterson is General Chair of CIBB17 (Cagliari, Italy), and session chair of Machine Learning for Enhancing Biomedical Data Analysis at IJCNN17. Dr. Peterson was General Chair of CIBB12, co-chair of CIBB09, and has chaired special sessions on neural networks at IJCNN10, IJCNN09, IJCNN08, IJCNN07, IJCNN06, ICMLA07, ICMLA08, CIBB12, CIBB09, CIBB11,CIBB10, CIBCB11, and served on the international program committees for PATTERNS11, INTELLI13, ICEIS12, HISB11, PReMI11, ICEIS11, ICAART10, ICEIS2010, MI-CAI09,ISMDS09, SoCPaR09, NaBIC09, ICAART09, BIOSIGNALS09, HEALTHINF09, ECAI08, ICEIS09, ISMDS08, ICEIS08, BIOSIGNALS08, HEALTHINF08, EUROCON07, FOCI07, CIBCB07, CIBCB06. He is a member of the Computational Intelligence Society of IEEE, member of the Task Force on Neural Networks (IEEE-CIS-BBTC), Houston-Area Chapter Chair (IEEE-CIS), member of IEEE-CIS-BBTC, Vice Chair for Finance (IEEE CS-TCCLS), and Editor-in-Chief of Source Code for Biology and Medicine.
Dr. Peterson’s research focuses on development of novel techniques for optimization based on neural adaptive learning with metaheuristics (neural networks, genetic algorithms, swarm intelligence, evolution strategies) and information retrieval using duo-mining (data and text), n-grams, and non-linear and linear dimensional reduction with manifold learning and eigendecomposition. Manifold learning techniques include kernel PCA, Laplacian eigenmaps, local linear embedding, locally preserving projections, neural gas, self-organizing maps, and Sammon maps. Random matrix theory involving noise removal techniques focus on the limiting spectral densities of noise eigenvalues based on the Marcenko-Pastur and Tracy-Widom laws. Other areas of focus include Monte Carlo simulation for uncertainty using skew-normal, Laplace, Levy-Stable, triangle, and beta probability distributions as well as fractal geometry and use of the Marcenko-Pastur law in quantitative finance for asset risk modeling of fat-tail distributions.
Patent Number: US2015377906, Dec 31 2015