Explore familywise error rate control for statistical testing in brain image ontologies in this 38-minute conference talk by Daniel Tward. Delve into the challenges of statistical neuroimaging, including false positives and complex brain geometry. Learn about brain atlases, spatial patterns, and the conversion of test statistics to p-values. Examine methods for addressing false discovery rates and familywise error rate control. Discover how to build statistical models and apply the EM algorithm to sample data. Gain insights from related research papers and a poster presentation on statistical testing at multiple scales in brain ontologies.
Familywise Error Rate Control for Statistical Testing in Brain Image Ontologies
Computational Genomics Summer Institute CGSI via YouTube
Overview
Syllabus
Introduction
Goal
Brain
Spatial Scale
Complex Geometry
Brain Atlases
Spatial Patterns
Statistical Neuro Imaging
Converting Test Statistics to P Values
Problems with Statistical Testing
False Positives
False Discovery Rate Correction
Familywise Error Rate Control
Exchangeability
Example
Brain Atlas
Two Problems
Simplify the Problem
Build a Statistical Model
Sample Data
Scatter Plot
EM Algorithm
Test Statistic
Poster Presentation
Summary
Taught by
Computational Genomics Summer Institute CGSI
