David Bryant Keator

picture of David Bryant Keator

Associate Research Professor, Psychiatry & Human Behavior
School of Medicine
Operations Director - Neuroscience Imaging Center, Psychiatry & Human Behavior
School of Medicine

B.S., University of California, Irvine, 1994, Biological Sciences
M.Engr., California State University, Long Beach, 2001, Computer Science
Doctor of Computer Science, University of California, Irvine, 2015, Machine Learning

Phone: (949) 824-7870
Email: dbkeator@uci.edu

University of California, Irvine
Neuroscience Imaging Center
Irvine Hall rm. 163
Mail Code: 3960
Irvine, CA 92697
Research Interests
neuroimaging, dementia, PET, MRI, machine learning
URL
Academic Distinctions
Phi Kappa Phi National Honor Society
Graduate Dean's List Award - CSULB Department of Engineering and Computer Science
Research Abstract
For over twenty years I have been an active researcher in the fields of data science and neuroimaging. My research over the past five years has been focused in three principle domains described in subsequent sections below. I work in a variety of inter-disciplinary teams, bringing a unique technical perspective to collaborative biomedical research groups.

(1) Developing models and techniques in knowledge representation, management, and retrieval for problems in medicine.

The field of neuroimaging, and biomedical sciences in general, has lacked the capability to reuse and combine existing data and replicate existing studies due, in part, to the lack of structured metadata, the lack of publically available, well-described, datasets, and the lack of techniques for maintaining un-ambiguous and complete records of computation. In light of these needs, I have both developed and contributed to a variety of data formats, database technologies, and knowledge representation models for research. In 2006-2012 as part of the inter-disciplinary Function Biomedical Informatics Research Network project (U24 RR21992 NIMH PI: Potkin; Role: Co-I, Neuroinformatics Chair) we developed XML-based formats and associated open-sourced tools for capturing structured experiment metadata and derived data and designed and developed the first federated data management system for neuroimaging whose main contributions were an extensible schema capable of storing new data types without changes to the underlying organizational structure and supportive of multi-site federated queries. Building upon the success of these tools for research, we expanded our focus in 2011-2012 and built systems to dynamically share research-ready brain imaging data across clinical centers to enable use in clinical settings. My work on these technologies resulted in a variety of publically-accessible data repositories for the sharing of domain-specific imaging data where we found that a more extensible technology was needed to accommodate rapidly changing data and to capture workflows and data provenance in an un-ambiguous, semantically annotated manner.

In 2012 I convened an open working group to develop a next-generation format capable of capturing data from all stages of the scientific research process, from raw data to derived results and provenance. The result of this working group is the Neuroimaging Data Model (NIDM; http://nidm.nidash.org), the first semantic web-enabled metadata format for neuroimaging. NIDM has been the topic of recent international workshops taught by myself and collaborators ("Understanding NIDM Scientific Interest Group", Kuala Lumpur, Malaysia 2017; "Using the neuroimaging data model (NIDM) for databasing and querying complex data" at the Japan Society for Neuroscience 2016 in Yokohama, Japan; "NIDM Workshop" at Neuroinformatics Congress 2015 in Cairns, Australia). NIDM is being incorporated into the most popular software packages for neuroimaging analyses, we are actively working on a Python package to support tool developers called PyNIDM, and is a main component of a collaborative grant with the University of Massachusetts to improve reproducibility in neuroimaging computation (1P41EB019936-01A1). Further, the NIDM technology has been used to support the multi-disciplinary UCI Conte Center demonstrating its breadth in representing raw and derived data across species and scientific disciplines. My future research interests include extending popular machine learning APIs (e.g. Python scikit-learn) to accommodate NIDM-formatted metadata, augment meta-analysis models to use the NIDM-Results model directly, and to record workflows from these systems according to the NIDM-Workflow model.

(2) The development of models for mining data in neuroimaging and medicine.

Functional brain imaging is a common tool in monitoring the progression of neurodegenerative and neurological disorders. I have made contributions to both improving data acquisition quality, in identifying functional brain imaging-derived features used to detect a variety of psychiatric and neurological problems, and in data mining and analysis methods. For example, in 2015 I developed a probabilistic graphical modeling approach to improve system tuning in Positron Emission Tomography (PET) that has outperformed manufacturer (Siemens Inc.) methods. These methods have been integrated into the UCI Neuroscience Imaging Center (NIC) to improve the resolution of acquired data for all NIC investigators. In 2014 I extended algorithms from the computer vision community to detect regional functional abnormalities in both Alzheimer's disease and traumatic brain injury (National Football League players) as assessed with functional neuroimaging. My results compared favorably with other published classification results that had been tuned for particular disorders and outperform those of a blinded expert human rater. Currently I am working on novel predictive algorithms for dementia and models for functional connectivity analysis.

(3) Predictive analytics for dementia prediction in Down's Syndrome (DS), traumatic brain injury (TBI), Schizophrenia, and Alzheimer's disease (AD).

Understanding the how various disorders such as TBI, Schizophrenia, and AD affect the brain as assessed with neuroimaging technologies is important to both individualized patient care and treatment development. Further, identifying biomarkers for early dementia prediction in DS and the aged population has promise to improve treatment development and outcomes. I have made significant technical contributions to studies evaluating how one quantifies neurodegeneration in AD. I made a significant contribution to the understanding of the brain areas impacted by concussion-related TBI associated with playing professional American football and designed an algorithm to automatically classify subjects with concussion-related TBI that yielded high accuracy. Further, I have made significant contributions to the understanding of how Schizophrenia affects neuroanatomical structure and function. More recently, I have been working on understanding dementia in DS and finding brain-based biomarkers for early dementia prediction (see invited presentation "Baseline 18F-AV-45 PET Predictors of Dementia Transition in Down's Syndrome", Alzheimer's and Parkinson's Disease Congress, Vienna Austria, March 25-31, 2017). This work, the first of its kind, shows that the spatial distribution of amyloid in the brain is critical for differentiating sub-types of patients with DS prone to early dementia using data acquired years prior to their clinical conversion. This work incorporates both survival models and predictive analytics using spatial patterns of amyloid.
Publications
Selected Publications:

Keator D.B., Chen J., Nichols N., Fana F., Stern H., Baram T.Z., Small S.L. A Semantic Cross-Species Derived Data Management Application. Data Science Journal. 2017 16, p.45. doi: http://doi.org/10.5334/dsj-2017-045

Ghosh SS, Poline JB, Keator DB, Halchenko YO, Thomas AG, Kessler DA, Kennedy DN. A very simple, re-executable neuroimaging publication. F1000Research. 2017 Feb 10;6.

Amen DG, Trujillo M, Keator D., Taylor DV, Willeumier K, Meysami S, Raji CA. Gender-Based Cerebral Perfusion Differences in 46,034 Functional Neuroimaging Scans. Journal of Alzheimer's Disease. 2017 Aug 4(Preprint):1-10.

Doran E, Keator D.B., Head E., Phelan M.J., Kim R., Totoiu M., Barrio J., Small G., Potkin S.G., Lott I. Down Syndrome, Partial Trisomy 21, and Absence of Alzheimer's Disease: The role of APP. Journal of Alzheimer's Disease. 56.2 (2017): 459-470.

Maumet C., Auer T., Bowring A., Chen G., Das S., Flandin G., Ghosh S., Glatard T., Gorgolewski K., Helmer K., Jenkinson M., Keator D.B., Nichols N, Poline J.B., Reynolds R., Sochat V., Turner J., Nichols T. Sharing brain mapping statistical results with the neuroimaging data model. Scientific Data. Scientific Data, 3, 160102 (2016). http://doi.org/10.1038/sdata.2016.102.

Gorgolewski K.J.,Auer T.,Calhoun V.,Craddock C.,Das S.,Duff E.,Flandin G.,Ghosh S.,Halchenko Y.,Handwerker D.,Hanke M.,Keator D.B.,Li X.,Maumet C.,Michael Z.,Nichols B.N.,Nichols T.,Poline J.B.,Roken A.,Schaefer G.,Sochat V.,Turner J.A.,Varoquaux G.,Poldrack R. The Brain Imaging Data Structure: a standard for organizing and describing outputs of neuroimaging experiments. Data Science Journal. 2016.

Miller RL, Vergara V, Keator DB, Calhoun VD. A Method for Inter-temporal Functional Domain Connectivity Analysis: Application to Schizophrenia Reveals Distorted Directional Information Flow. IEEE Trans Biomed Eng. August 16, 2016.

Keator D.B., Van Erp T.G, Glover G.H., Mueller B.A., Turner J.A., Liu T., Greve D., Voyvodic J., Rasmussen J., Brown G., Calhoun V.D., Lee H., Ford J., Mathalon D., Diaz M., O'Leary D., Gadde S., Preda A., Lim K., Wible C., Stern H., Belger A., McCarthy G., Ozyurt B., Potkin S.G., FBIRN. The Function Biomedical Informatics Research Network Data Repository. Neuroimage Special Issue on Brain Imaging Repositories. 124: 1074-1079.

Ambite, J. L., Tallis, M., Alpert, K., Keator, D. B., King, M., Landis, D., Konstantinidis, G., Calhoun, V. D., Potkin, S. G., Turner, J. A., & Wang, L. (2015) SchizConnect: Virtual Data Integration in Neuroimaging. Data Integr Life Sci, 9162:37-51.

Keator, D.B., Ihler, A. Crystal Identification in Positron Emission Tomography Using Probabilistic Graphical Models. Nuclear Science, IEEE Transactions on. 62.5 (2015): 2102-2112.

Wang L., Alpert K., Calhoun V., Keator D.B., King M., Kogan A., Landis D., Tallis M., Potkin S.G., Turner J.A., Amite J.L. SchizConnect: Mediating Schizophrenia Neuroimaging Databases for Large-Scale Integration. Neuroimage Special Issue on Brain Imaging Repositories. 2015.

Lee, H.J., Preda A., Ford J.M., Mathalon D.H., Keator D.B., Van Erp T.G.M., Turner J.A., Potkin S.G.. "Functional Magnetic Resonance Imaging of Motor Cortex Activation in Schizophrenia." Journal of Korean medical science. 2015; 30, no. 5: 625-631.

Van Erp TGM, Stark C., Rasmussen J., Turner J., Calhoun V., Razzak S., Lim K.O., Mueller B., Brown G., Bustillo J., Vaidya J., McEwen S., Voyvodic J., Belger A., Mathalon D., Keator D.B., Preda A., Nguyen D., Ford J., Potkin S.G.. "Hippocampal Subfield Volume Abnormalities in Individuals with Schizophrenia." Neuropsychopharmacology, 2014; vol. 39.

Van Erp TG, Greve DN, Rasmussen J, Turner J, Calhoun VD, Young S, Mueller B, Brown GG, McCarthy G, Glover GH, Lim KO, Bustillo JR, Belger A, McEwen S, Voyvodic J, Mathalon DH, Keator D, Preda A, Nguyen D, Ford JM, Potkin SG, Fbirn. A multi-scanner study of subcortical brain volume abnormalities in schizophrenia. Psychiatry Res. 2014;222(1-2):10-6.

Rafii M.S., Baumann T.L., Bakay R.A.E, Ostrove J.M., Siffert J., Fleisher A.S., Herzog C.D., Barba D., Pay M., Tuszynski M.H., Salmon D., Kordower J.H., Bishop K., Keator D.B., Potkin S.G., Bartus R.T. A phase 1 study of sterotactic gene delivery of AAV2-NGF for Alzheimer's disease. Alzheimer's Disease & Dementia 2014 Jan 7.

Potkin, S.G., Keator D.B., Kesler-West M.L., Nguyen D.D., VanErp T.G.M., Mukherjee J., Shah N., Preda A. D2 Receptor Occupancy Following Lurasidone Treatment in Patients with Schizophrenia or Schizoaffective Disorder. CNS Spectrums, 2013 Sep 30:1-6.

Keator D.B., Helmer K., Steffener J., Turner J.A., Van Erp T.G.M., Gadde S., Ashish N., Burns G.A., Nichols B.N. Towards structured sharing of raw and derived neuroimaging data across existing resources. Neuroimage. 2013 Nov 15;82:647-61

Keator, D.B. "Modality Neutral Techniques for Brain Image Understanding." Machine Learning and Interpretation in Neuroimaging. 2012: 84-92.

Keator D.B., Fallon J.H., Lakatos A., Fowlkes C., Potkin S.G., Ihler A. Feed-Forward Hierarchical Model of the Ventral Visual Stream Applied to Functional Brain Image Classification. Journal of Human Brain Mapping. 2012 Jul 30.

Chervenak A.L., van Erp T.G., Kesselman C., D'Arcy M., Sobell J., Keator D., Dahm L., Murray J., Law M., Hasso A., Ames J., Macciardi F., Potkin S.G. A system architecture for sharing de-identified, research-ready brain scans and health information across clinical imaging centers. Studies in Health Technology and Informatics. 2012; 175:19-28.

Poline J.B., Breeze J., Ghosh S., Gorgolewski K., Halchenko Y., Hanke M., Haselgrove C., Helmer K., Keator D.B., Marcus D., Poldrack R., Schwartz Y., Ashburner A., Kennedy D. Data sharing in neuroimaging research. Frontiers in Neuroinformatics. 2012; 6:9.

Glover G.H., Mueller B.A., Turner J.A., Van Erp T.G, Liu T., Greve D., Voyvodic J., Rasmussen J., Brown G., Keator D.B., Calhoun V.D., Lee H., Ford J., Mathalon D., Diaz M., O'Leary D., Gadde S., Preda A., Lim K., Wible C., Stern H., Belger A., McCarthy G., Ozyurt B., Potkin S.G., FBIRN. Function Biomedical Informatics Research Network Recommendations for Prospective Multi-Center Functional Magnetic Resonance Imaging Studies. Journal of Magnetic Resonance Imaging. 2012 Feb 7.

Rasmussen J., Lakatos A., Van Erp T., Kruggel F., Keator D.B., Fallon J., Potkin S., Alzheimer's Disease Neuroimaging Initiative. Empirical derivation of the denominator region for computing degeneration sensitive 18fluorodeoxyglucose ratios in Alzheimer's Disease based on the ADNI study. Biochim Biophy Acta - Molecular Basis of Disease. 2012 Mar;1822(3):457-66.

Van Erp, TGM, Chervenak A., Kesselman C., D'Arcy M., Sobell J., Keator D., Dahm L. et al. "Infrastructure for sharing standardized clinical brain scans across hospitals." In Bioinformatics and Biomedicine Workshops (BIBMW), 2011 IEEE International Conference on, pp. 1026-1028. IEEE, 2011.

Helmer KG, Ambite JL, Ames J, Ananthakrishnan R, Burns G, Chervenak AL, Foster I, Liming L, Keator D, Macciardi F, Madduri R, Navarro JP, Potkin S, Rosen B, Ruffins S, Schuler R, Turner JA, Toga A, Williams C, Kesselman C; for the Biomedical Informatics Research Network. Enabling collaborative research using the Biomedical Informatics Research Network (BIRN). J Am Med Inform Assoc. 2011 Apr 22.

Borghammer P, Hansen SB, Eggers C, Chakravarty MM, Vang K, Aanerud JF, Hilker R, Heiss WD, Rodell A, Munk OL, Keator D, and Gjedde A. Glucose metabolism in small subcortical structures in Parkinson's disease. Acta Neurol Scand. 2011.

Gadde S., Aucoin N., Grethe J.S., Keator D.B., Marcus D.S., Pieper S., FBIRN, MBIRN, BIRN-CC. XCEDE: An Extensible Schema for Biomedical Data. Neuroinformatics. 2011 Apr 9.

Amen DG, Newberg A, Thatcher R, Jin Y, Wu J, Keator D, Willeumier K. Impact of playing American professional football on long-term brain function. J Neuropsychiatry Clin Neurosci. 2011 Fall;23(1):98-106.

Ozyurt I.B., Keator D., Wei D., Fennema-Notestine C., Pease K., Bockholt B., Grethe J. Federated Web-accessible Clinical Data Management within an Extensible NeuroImaging Database. Neuroinformatics. 2010;23(1):98-106.

Lakatos A., Derbeneva O., Younes D., Keator D.B., Bakken T., Lvova M., Brandon M., Guffanti G., Reglodi D., Saykin A., Weiner M., Macciardi F., Schork N., Wallace D., Potkin S., ADNI. Association between mitochondrial DNA variations and Alzheimer's Disease in the ADNI cohort. Neurobiology of Aging. 2010;31(8):1355-63.

Keator, D.B., Wei, D., Gadde, S., Bockholt, H., Grethe, J.S., Marcus, D., Aucoin, N., Ozyurt, B. Derived Data Storage and Exchange Workflow for Large-Scale Neuroimaging Analyses on the BIRN Grid. Front Neuroinformatics. 2009;3:30.

Keator D.B., Mukherjee J., Preda A., Highum D., Lakatos A., Gage A., Potkin S.G.. "Dopamine D2 and D3 receptor occupancy of cariprazine in schizophrenic patients." Schizophrenia Bulletin, 2009; vol. 35, pp. 154-154.

Fallon, J., Keator, D.B. Commentary on "In silico modeling system: a national research resource for simulation of complex brain disorders.". Alzheimer's Dementia. 2009; Jan: 5(1):5-6.

Abe, S., Preda A., Turner J., Keator D.B., Potkin S.G., and F. Birn. "DTI Co-registration Method Comparison Based on DTI Tractography Analysis in the Human Brain." Neuroimage 47 (2009): S51.

Potkin, S.G., Turner, J.A., Guffanti, G., Lakatos, A., Torri, F., Keator D.B., Macciardi F. Genome-wide strategies for discovering genetic influences on cognition and cognitivie disorders: methodological considerations. Cognitive Neuropsychiatry. 2009;Jul;14(4):391-418.

Potkin, S.G., Turner, J.A., Fallon, J.A., Keator, D.B., Guffanti, G., Macciardi, F., FBIRN. Gene Discovery Through Imaging Genetics - Identification of Two Novel Genes Associated with Schizophrenia. Molecular Psychiatry. 2009 Apr;14(4):416-28.

Ford, J.; Roach, B.; Turner, J.; Brown, G.; Greve, D.; Wible, C.; McCarthy, G.; Lauriello, J.; Belger, A.; Mueller, G.; Calhoun, V.; Preda, A.; Keator, D.; O'Leary, D.; Lim, K.; Glover, G.; Potkin, S.; Mathalon, D. Tuning in to the voices: A multi-site fMRI study of auditory hallucinations. Schizophrenia Bulletin. 2009 Jan;35(1):58-66.

Keator, D.; Grethe, J.S.; Marcus, D.; Ozyurt, B.; Gadde, S.; Murphy, S.; Pieper, S.; Greve, D.; Notestine, R.; Bockholt, H.J; Papadopoulos, P.; Function BIRN; Morphometry BIRN; BIRNCoordinating Center. A National Human Neuroimaging Collaboratory Enabled By The Biomedical Informatics Research Network (BIRN). IEEE Transactions on Information Technology in Biomedicine. 2008 Mar;12(2):162-72.
Grants
NIBIB 5P41EB019936-02 Sub-Award PI: WA00433491 Center for Reproducible Neuroimaging Computation (CRNC) The Center for Reproducible Neuroimaging Computation (CRNC), a Biomedical Technology Research Center, is dedicated to establishing and promoting reproducible neuroimaging research through the use of semantic web and software container technologies.
NIA 5U01AG051412-03 Sub-Award Co-I: 3U01AG051412-03S1 Biomarkers of Alzheimer's Disease in Down Syndrome Goals: The overall aim of this project is to identify biomarkers associated with the progression of Alzheimer's disease (AD) from its prodromal stages to frank dementia in adults with Down syndrome (DS).
NIMH R01 MH108155 (PI: Rao), Co-I: Effects of Childhood Maltreatment on Neurocircuitry in Adolescent Depression
NIDA R01 DA040966 (PI: Rao), Co-I: Prevention of Adolescent Risky Behaviors: Neural Markers of Intervention Effects
NIMH 1P50MH096889 (PI: Baram), Co-I: Fragmented Early Life Environmental and Emotional / Cognitive Vulnerabilities
Professional Societies
IEEE
Organization of Human Brain Mapping
Graduate Programs
Biomedical Engineering

Last updated
01/30/2018