Welcome to NEUROFINDER
A common problem in the analysis of neuroscience imaging data is source extraction: identifying neurons from time-varying image sequences. There are many existing algorithms but they have yet to be compared. The goal of NEUROFINDER is to compare these algorithms in a common environment on standardized, vetted data sets, and begin to integrate algorithms into a common platform. The notebooks below show how to load the NEUROFINDER data, and how to run algorithms and evaluate their performance.
For more information, visit the project page or come chat with us.
Tutorials
Evaluating an algorithm (block-based)
Evaluating an algorithm (feature-based)
Writing an algorithm (using Spark/Thunder)
Writing an algorithm (using pure scientific Python)
Return to welcome