Other formats:
BibTeX
LaTeX
RIS
@article{1305888, author = {Janoušová, Eva and Schwarz, Daniel and Kašpárek, Tomáš}, article_location = {CLARE}, article_number = {3}, doi = {http://dx.doi.org/10.1016/j.pscychresns.2015.03.004}, keywords = {Computational neuroanatomy; Classification; Intersubject principal component analysis (isPCA); Modified maximum uncertainty linear; discriminant analysis (mMLDA); Centroid method; Average linkage}, language = {eng}, issn = {0925-4927}, journal = {Psychiatry Research: Neuroimaging}, title = {Combining various types of classifiers and features extracted from magnetic resonance imaging data in schizophrenia recognition}, volume = {232}, year = {2015} }
TY - JOUR ID - 1305888 AU - Janoušová, Eva - Schwarz, Daniel - Kašpárek, Tomáš PY - 2015 TI - Combining various types of classifiers and features extracted from magnetic resonance imaging data in schizophrenia recognition JF - Psychiatry Research: Neuroimaging VL - 232 IS - 3 SP - 237-249 EP - 237-249 PB - ELSEVIER IRELAND LTD SN - 09254927 KW - Computational neuroanatomy KW - Classification KW - Intersubject principal component analysis (isPCA) KW - Modified maximum uncertainty linear KW - discriminant analysis (mMLDA) KW - Centroid method KW - Average linkage N2 - We investigated a combination of three classification algorithms, namely the modified maximum uncertainty linear discriminant analysis (mMLDA), the centroid method, and the average linkage, with three types of features extracted from three-dimensional T1-weighted magnetic resonance (MR) brain images, specifically MR intensities, grey matter densities, and local deformations for distinguishing 49 first episode schizophrenia male patients from 49 healthy male subjects. The feature sets were reduced using intersubject principal component analysis before classification. By combining the classifiers, we were able to obtain slightly improved results when compared with single classifiers. The best classification performance (81.6% accuracy, 75.5% sensitivity, and 87.8% specificity) was significantly better than classification by chance. We also showed that classifiers based on features calculated using more computation-intensive image preprocessing perform better; mMLDA with classification boundary calculated as weighted mean discriminative scores of the groups had improved sensitivity but similar accuracy compared to the original MLDA; reducing a number of eigenvectors during data reduction did not always lead to higher classification accuracy, since noise as well as the signal important for classification were removed. Our findings provide important information for schizophrenia research and may improve accuracy of computer aided diagnostics of neuropsychiatric diseases. ER -
JANOUŠOVÁ, Eva, Daniel SCHWARZ and Tomáš KAŠPÁREK. Combining various types of classifiers and features extracted from magnetic resonance imaging data in schizophrenia recognition. \textit{Psychiatry Research: Neuroimaging}. CLARE: ELSEVIER IRELAND LTD, 2015, vol.~232, No~3, p.~237-249. ISSN~0925-4927. Available from: https://dx.doi.org/10.1016/j.pscychresns.2015.03.004.
|