Title | ||
---|---|---|
Detecting Regions of Interest in fMRI: An Application on Exploratory-based Data Analysis |
Abstract | ||
---|---|---|
Cluster validity has been mainly used to evaluate the quality of individual clusters, and compare whole partitions resulting from dierent or same (using dierent parameters) clustering algo- rithms (13). However, depending on the applica- tion, the demands for a validity measure may dif- fer, inducing the necessity of introducing new mea- sures which will suit to the problem under investiga- tion. We introduce a new statistical concept for quan- titative validation of fMRI analysis methods based on exploratory data analysis (EDA) algorithms. A "greedy" algorithm which combines partitions on a sequential basis is applied on 100 runs of a cluster- ing algorithm with randomized cluster initialization. As a result, a fuzzy partition is derived where every data point is assigned to a cluster with a certain de- gree of "support". The purpose of this measure is to check the stability of the resulting clustering struc- tures and improve the diagnostic reliability of fMRI post-processing strategies. |
Year | DOI | Venue |
---|---|---|
2002 | 10.1109/FUZZ.2002.1006726 | World Congress on Computational Intelligence |
Keywords | DocType | Volume |
biomedical MRI,data analysis,feature extraction,fuzzy set theory,pattern clustering,cluster validity,exploratory data analysis,fMRI,functional magnetic resonance imaging,fuzzy partition,greedy algorithm,medical image processing,regions of interest detection | Conference | 2 |
Citations | PageRank | References |
0 | 0.34 | 7 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Evgenia Dimitriadou | 1 | 136 | 9.32 |
Markus Barth | 2 | 195 | 20.94 |
Christian Windischberger | 3 | 381 | 37.71 |
K. Hornik | 4 | 633 | 209.19 |
Ewald Moser | 5 | 134 | 20.53 |