*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Erik Reinertsen
PhD Defense
Date: Tuesday, Aug 15, 2017
Time: 10 - 11 am
Location: Emory University School of Medicine room 170A
Committee members:
- Gari Clifford, DPhil (advisor)
- Shamim Nemati, PhD
- Amit Shah, MD, MSCR
- Eberhard Voit, PhD
- Lee Cooper, PhD
Title: Dichotomizing illness from cardiovascular and activity time series
Abstract:
Digital sensors on smartphones and wearables measure huge quantities of time series data. Patterns in these data - such as heart rate (HR) and locomotor activity - reflect changes in physiology and behavior associated with mental and/or cardiovascular illness. However, most feature extraction approaches do not capture how noise and information varies with time scale, or is transferred between different signals. This thesis explores the use of time domain, frequency domain, and complexity measures over several time scales to train machine learning algorithms to classify illness. First, a novel segmentation method based on periods of minimum heart rate is shown to improve the classification of post-traumatic stress disorder (PTSD) from healthy controls using heart rate variability metrics. Second, the relationship between analysis window length and classifier accuracy is evaluated in the context of using HR and activity features to dichotomize subjects with schizophrenia from healthy controls. Third, multiscale transfer entropy and network theoretical metrics of heart rate and activity are evaluated as complexity measures that contribute to the classification of atrial fibrillation and PSTD. This work contributes to the growing field of objective physiological sensing that could enable long-term ambulatory monitoring of clinically significant conditions.
-----------------------------------
To join the meeting on a computer or mobile phone : https://bluejeans.com/383674918
Here are the details for the meeting you scheduled:
-----------------------------------
Connecting directly from a room system?
1) Dial: 199.48.152.152 or bjn.vc
2) Enter Meeting ID: 383674918
Just want to dial in on your phone?
1) +1.408.740.7256 (United States)
+1.888.240.2560 (US Toll Free)
+1.408.317.9253 (Alternate number)
(http://bluejeans.com/numbers)
2) Enter Meeting ID: 383674918
3) Press #
