Sciweavers

BIBE
2008
IEEE

Feature selection and classification for assessment of chronic stroke impairment

13 years 5 months ago
Feature selection and classification for assessment of chronic stroke impairment
Recent advances of robotic/mechanical devices enable us to measure a subject's performance in an objective and precise manner. The main issue of using such devices is how to represent huge experimental data compactly in order to analyze and compare them with clinical data efficiently. In this paper, we choose a subset of features from real-time experimental data and build a classifier model to assess stroke patients' upper limb functionality. We compare our model with combinations of different classifiers and ensemble schemes, showing that it outperforms competitors. We also demonstrate that our results from experimental data are consistent with clinical information, and can capture changes of upper-limb functionality over time.
Jae-Yoon Jung, Janice I. Glasgow, Stephen H. Scott
Added 12 Oct 2010
Updated 12 Oct 2010
Type Conference
Year 2008
Where BIBE
Authors Jae-Yoon Jung, Janice I. Glasgow, Stephen H. Scott
Comments (0)