EE440 Brain-Machine Learning

Catalog Data & Credits (Recitation-Lab-Total)

EE440 Brain-Machine Learning (2-2-3): This course develops the concepts from machine learning, signal processing, and neuroscience required to understand how modern brain-machine interfaces - technologies that interact with the nervous system for therapeutic or rehabilitative purposes - interpret and respond to brain signals. Examples of these technologies include retinal prostheses for the blind and brain-driven limb protheses for amputees. The course also examines brain-machine interfaces at the system level through directed readings of the scientific and engineering literature. Specific course topics include basic neuroanatomy and neurophysiology for engineers and statistical techniques for the dimensionality reduction, de-noising, classification, and clustering of neural signals.

Pre-requisites

Course Objectives

1. Apply fundamental anatomical and physiological neuroscience principles to identify the engineering problems solved by modern brain-machine interfaces (BMIs) and understand their design and operation.
2. Describe machine learning and signal processing algorithms commonly used in BMIs.
3. Apply neuroscience and machine learning knowledge to the analysis of real neurophysiological data using Matlab.
4. Critically interpret and present research and review papers from the scientific and engineering literature.
5. Discuss the state-of-the art in commercial, clinical, and research BMIs.