VSP - Collaboration on Brain Imaging and Behavioral Data Analysis

Abstract

Visit the Lifelong Brain and Cognition Lab at Beckman Institute and pursue mutual benefits on brain and behavioural data analysis. The lab has been collecting multi- modality brain and behavioural data from young and older adults taking part in longitudinal physical, cognitive and nutritional interventions. While mining their rich sets of empirical data requires increasingly sophisticated data analysis methods, the lab personnel there with non-engineering majors are only familiar with traditional and simple analysis methods. They indeed are very interested in the pattern analysis and Bayesian network frameworks I have been developing for different MR modalities. My cutting-edge data analysis techniques will give a profound impact on their cognitive research by collaborating with their expertise in psychology and neuroscience. The analysis pipelines unavoidably require many user-decided inputs such as cost functions, kernels, feature selections, regularization terms, regions of interest, hemodynamic response models, group network inference approaches, etc. To make matters worse, there arenÕt robust routines which can help the users decide quickly how to initialize these parameters. Instead, they make many failed attempts which could be reduced by the proper skill sets, including catching typical error characteristics in the outputs. Hence, it is crucial for me to collaborate with them in person and to produce many possible outputs they might get when running the pipelines on their data sets. It will take at least three weeks to ensure that the future applications of the pipelines to their data sets will produce valid results.

Document Details

Document Type
DoD Grant Award
Publication Date
Jun 10, 2016
Source ID
N629091612037

Entities

People

  • Loan Vo

Organizations

  • Office of Naval Research
  • United States Navy

Tags

Readers

  • Gulf War Illness and Chronic Multisymptom Illness in Veterans.
  • Neural Network Machine Learning.
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks