Navigation

Ravi Bansal, PhD

Investigator, Research on Children, Youth and Families
Professor of Research Pediatrics, Keck School of Medicine of USC

Ravi Bansal, PhD, is an expert in signal and medical image processing, including anatomical MRI, Diffusion Tensor Images, functional MRI, and Arterial Spin Labeling. He has published more than 100 peer reviewed papers in the development and application of cutting-edge imaging processing technologies to the study of normal brain maturation and developmental psychopathologies. These methodological include mathematical and statistical methods for making computer-based neuropsychiatric diagnoses using MR images of the brain, improving tissue definition while accounting for variations in tissue contrast that are caused by variation in tissue water content, automating shape analyses of brain regions delineated on high-resolution anatomical MR images, and applying these analytic tools to defining the biological bases for neuropsychiatric disorders. He has developed machine learning algorithms for diagnosing an individual as having one or the other neuropsychiatric disorder, or has having illness or not using MRI data of the brain. These algorithms used brain measures defined as the scaling coefficients computed by spherical wavelet analyses and applied a method for hierarchical clustering for learning the classification rule and for diagnosing a new individual. Bansal and his team rigorously evaluated the performance of the machine learning algorithms using split half cross validation. Additionally, he has applied the mathematical and statistical methods that he developed to study the brain-based effects of environmental neurotoxicants.

Education

Graduate School

Northeastern University, Bostom, Massachusettes, M.S (Electrical Engineering), 1994 

Yale University, New Haven, Connecticut, M.Phil. and Ph.D. (Electrical Engineering), Honor, 1996 & 1999

Publications

Serotonin signaling modulates the effects of familial risk for depression on cortical thickness.Bansal R, Peterson BS, Gingrich J, Hao X, Odgerel Z, Warner V, Wickramaratne PJ, Talati A, Ansorge M, Brown AS, Sourander A, Weissman MM. Psychiatry Res. 2016 Feb 28;248:83-93. doi: 10.1016/j.pscychresns.2016.01.004. Epub 2016 Jan 6.

Identification of a circuit-based endophenotype for familial depression. Dubin MJ, Weissman MM, Xu D, Bansal R, Hao X, Liu J, Warner V, Peterson BS. Psychiatry Res. 2012 Mar 31;201(3):175-81. doi: 10.1016/j.pscychresns.2011.11.007. Epub 2012 Apr 18. Erratum in: Psychiatry Res. 2013 Jan 30;211(1):92. 

Relationship of resting EEG with anatomical MRI measures in individuals at high and low risk for depression. Bruder GE, Bansal R, Tenke CE, Liu J, Hao X, Warner V, Peterson BS, Weissman MM. Hum Brain Mapp. 2012 Jun;33(6):1325-33. doi: 10.1002/hbm.21284. Epub 2011 Apr 15.

Neuroanatomical correlates of intellectual ability across the life span. Goh S, Bansal R, Xu D, Hao X, Liu J, Peterson BS. Dev Cogn Neurosci. 2011 Jul;1(3):305-12. doi: 10.1016/j.dcn.2011.03.001. Epub 2011 Mar 17.

Cerebellar morphology in Tourette syndrome and obsessive-compulsive disorder. Tobe RH, Bansal R, Xu D, Hao X, Liu J, Sanchez J, Peterson BS. Ann Neurol. 2010 Apr;67(4):479-87. doi: 10.1002/ana.21918.

PubMed Link
PubMed Link - Ravi Bansal

Research

  • Signal and medical image processing
    • Anatomical MRI
    • Diffusion Tensor Images
    • Functional MRI
    • Arterial Spin Labeling

Ravi Bansal, PhD, is an expert in signal and medical image processing, including anatomical MRI, Diffusion Tensor Images, functional MRI, and Arterial Spin Labeling. He has published more than 100 peer reviewed papers in the development and application of cutting-edge imaging processing technologies to the study of normal brain maturation and developmental psychopathologies. These methodological include mathematical and statistical methods for making computer-based neuropsychiatric diagnoses using MR images of the brain, improving tissue definition while accounting for variations in tissue contrast that are caused by variation in tissue water content, automating shape analyses of brain regions delineated on high-resolution anatomical MR images, and applying these analytic tools to defining the biological bases for neuropsychiatric disorders. He has developed machine learning algorithms for diagnosing an individual as having one or the other neuropsychiatric disorder, or has having illness or not using MRI data of the brain. These algorithms used brain measures defined as the scaling coefficients computed by spherical wavelet analyses and applied a method for hierarchical clustering for learning the classification rule and for diagnosing a new individual. Bansal and his team rigorously evaluated the performance of the machine learning algorithms using split half cross validation. Additionally, he has applied the mathematical and statistical methods that he developed to study the brain-based effects of environmental neurotoxicants.