Units:

12

Description:

This course is motivated by the increasing societal need to address healthcare problems through Big Data analysis and design of novel sensing modalities. All of the neuroscience background will be introduced in the class. Students are expected to have a strong background in linear algebra and probability (18-290, 36-217 or equivalent, senior or graduate standing preferred; discuss suitability of background with the instructor). The course will introduce students to concepts in data-science through hands-on data analyses oriented towards diagnosing and treating neural disorders such as epilepsy, traumatic brain injuries, stroke, etc. The eventual goal of the course is to produce a workforce of engineers and data-scientists who can work closely with clinicians to improve healthcare of tomorrow. We will discuss various modalities that are used to acquire clinical data, both invasively and noninvasively, and challenges in design, instrumentation, and data analyses for new modalities and multi-modal data. By the end of the course, the students will acquire statistical tools for modern data analytics and algorithms for efficient analyses such as optimization for imaging and inference, and machine-learning techniques. The analytics will be presented and performed with a deep understanding of the underlying physics of neural signal propagation as well as the particular disorder. These techniques will be used to image and infer neural activity as well as diagnose neural disorders on real and simulated data. Course projects will enable students to explore problems aimed at clinically-oriented research.

Prerequisites: 18-290, 36-217 or equivalent, senior or graduate standing preferred; discuss suitability of background with the instructor

Last Modified: 2017-11-16 7:01AM

Semesters offered:

• Fall 2017
• Spring 2017