Detailed Information on Publication Record
2023
State analysis of fMRI in amnestic mild cognitive impairment
GAJDOŠ, Martin, Marie NOVÁKOVÁ, Martin LAMOŠ, Pavel ŘÍHA, Irena REKTOROVÁ et. al.Basic information
Original name
State analysis of fMRI in amnestic mild cognitive impairment
Authors
GAJDOŠ, Martin (203 Czech Republic, guarantor, belonging to the institution), Marie NOVÁKOVÁ (203 Czech Republic, belonging to the institution), Martin LAMOŠ (203 Czech Republic, belonging to the institution), Pavel ŘÍHA (203 Czech Republic, belonging to the institution), Irena REKTOROVÁ (203 Czech Republic, belonging to the institution) and Michal MIKL (203 Czech Republic, belonging to the institution)
Edition
Czech – Austrian Workshop on Magnetic Resonance Imaging and Spectroscopy, 2023, Znojmo, 2023
Other information
Language
English
Type of outcome
Konferenční abstrakt
Field of Study
30103 Neurosciences
Country of publisher
Czech Republic
Confidentiality degree
není předmětem státního či obchodního tajemství
References:
Organization unit
Central European Institute of Technology
Keywords in English
Amnestic mild cognitive impairment; dynamic functional connectivity; sliding window analysis;
Tags
Reviewed
Změněno: 22/8/2024 14:19, Mgr. Eva Dubská
Abstract
V originále
Amnestic mild cognitive impairment (aMCI) is transitional state between normal aging and early dementia. In this work, we studied dynamic functional connectivity captured with sliding window analysis, whoch was performed on a dataset of 76 subjects (38 HC + 38 aMCI). We found significant differences in coverage in 2 out of 4 identified states. Moreover, with support vector machine, we were able to discriminate between these two groups with approx. 95% accuracy. In future work, we plan to crossvalidate presented classifier. Amnestic mild cognitive impairment (aMCI) is transitional state between normal aging and early dementia. In this work, we studied dynamic functional connectivity captured with sliding window analysis, whoch was performed on a dataset of 76 subjects (38 HC + 38 aMCI). We found significant differences in coverage in 2 out of 4 identified states. Moreover, with support vector machine, we were able to discriminate between these two groups with approx. 95% accuracy. In future work, we plan to crossvalidate presented classifier.
Links
NU21J-04-00077, research and development project |
| ||
90250, large research infrastructures |
|