Selection of a stroke risk model based on transcranial Doppler ultrasound velocity
DSpace at IIT Bombay
View Archive InfoField | Value | |
Title |
Selection of a stroke risk model based on transcranial Doppler ultrasound velocity
|
|
Creator |
MUKHOPADHYAY, S
DAS, I DAS, K |
|
Subject |
Dantzig selector
Gibbs sampling optimization parallel genetic algorithm sickle cell disease stochastic search variable selection SICKLE-CELL-DISEASE PARALLEL GENETIC ALGORITHM PULMONARY-HYPERTENSION VARIABLE SELECTION SUPERSATURATED DESIGNS DANTZIG SELECTOR ANEMIA PREVENTION ULTRASONOGRAPHY TRANSFUSIONS |
|
Description |
Increased transcranial Doppler ultrasound (TCD) velocity is an indicator of cerebral infarction in children with sickle cell disease (SCD). In this article, the parallel genetic algorithm ( PGA) is used to select a stroke risk model with TCD velocity as the response variable. Development of such a stroke risk model leads to the identification of children with SCD who are at a higher risk of stroke and their treatment in the early stages. Using blood velocity data from SCD patients, it is shown that the PGA is an easy-to-use computationally variable selection tool. The results of the PGA are also compared with those obtained from the stochastic search variable selection method, the Dantzig selector and conventional techniques such as stepwise selection and best subset selection.
|
|
Publisher |
TAYLOR & FRANCIS LTD
|
|
Date |
2014-10-15T15:08:09Z
2014-10-15T15:08:09Z 2012 |
|
Type |
Article
|
|
Identifier |
JOURNAL OF APPLIED STATISTICS, 39(12)2699-2712
http://dx.doi.org/10.1080/02664763.2012.725463 http://dspace.library.iitb.ac.in/jspui/handle/100/15119 |
|
Language |
en
|
|