omarcr / bcs-mit-computational-tutorial

Neural decoding of spike trains and local field potentials with machine learning in python

Speaker: Omar Costilla Reyes, PhD

Computation Tutorial

April 2nd, 2019 2:00pm - 4:00pm, McGovern Seminar Room (46-3189), Building 46 - MIT

Description: Neural decoding has applications in neuroscience from understanding neural populations to build brain-computer interfaces. In this computational tutorial, I will introduce neural decoding principles from a machine learning perspective using the Python programming language. The tutorial will be focused on data preprocessing, model selection and optimization for decoding neural information from spike trains and local field potentials. The studied dataset contains neural information from six cortical areas of the macaque brain spanning from the frontal to the occipital lobe.

Live tutorial on Youtube:

https://youtu.be/HDk1hczPky4

Download LFP data:

Mirror 1

Mirror 2

Download spike data:

Mirror 1

Mirror 2

Additional information

To follow along with the tutorial, bring a laptop with Anaconda Python 3.7 installed. Download for Windows/MacOS/Linux: https://www.anaconda.com/distribution/.

If you are using your own Python distribution or older versions of Anaconda:

• Update scikit-learn to version 0.20.3: pip install scikit-learn --upgrade
• Update matplotlib to version 3.0.3: pip install matplotlib --upgrade

Questions, feedback: @konet

Licence: BSD clause 3

Made with ❤️ @MIT