Premium
IC‐P1‐061: Bezoxazol derivatives containing trifluoromethoxy‐benzyl amino group for amyloid detection in APP transgenic mice using 19F magnetic resonance imaging
Author(s) -
Taguchi Hiroyasu,
Amatsubo Tomone,
Morikawa Shigahiro,
Matsuda Keiko,
Shirai Nobuaki,
Hirao Koichi,
Kato Masanari,
Morino Hisaya,
Kimura Hirohiko,
Nakano Ichiro,
Yoshida Chikako,
Okada Takashi,
Sano Mitsuo,
Tooyama Ikuo
Publication year - 2008
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2008.05.2503
Subject(s) - benzoxazole , chemistry , senile plaques , genetically modified mouse , magnetic resonance imaging , nuclear magnetic resonance , transgene , biochemistry , pathology , medicine , alzheimer's disease , organic chemistry , disease , physics , radiology , gene
the application of support vector machines (SVM) to MRI for detection of a variety of disease states. To date, the application of SVM to structural MR scans for the purpose of AD diagnosis has not been demonstrated using pathologically confirmed cases for training data, nor to differentiate different forms of dementia. The aims of this study were to assess how successfully SVMs assigned individual diagnoses and to determine whether data-sets combined from multiple scanners and different centers could be used to obtain effective classification of scans. Methods: We used linear support vector machines to classify the grey matter segment of T1-weighted MR scans from pathologically proven AD patients and cognitively normal elderly individuals obtained from two centers with different scanning equipment. Furthermore, we sought to use these methods to differentiate scans between patients suffering from AD from those with pathologically proven frontotemporal lobar degeneration (FTLD). Results: Up to 96% of AD patients were correctly classified using whole brain grey matter images. Data from different centers were successfully combined achieving comparable results from the separate analyses. Importantly, data from one center could be used to train a support vector machine to accurately differentiate AD and normal aging scans obtained from another center with different subjects and different scanner equipment. Our method correctly assigned 89% of patients with post-mortem confirmed diagnosis of either AD or FTLD to their respective group. Conclusions: Support vector machines successfully separate patients with AD from healthy aging subjects. They also perform well in the differential diagnosis of two different forms of dementia. The method is robust and can be generalized across different centers. This suggests an important role for computer based diagnostic image analysis for clinical practice.