TY - GEN
T1 - Empowering cortical thickness measures in clinical diagnosis of Alzheimer's disease with spherical sparse coding
AU - Zhang, Jie
AU - Fan, Yonghui
AU - Li, Qingyang
AU - Thompson, Paul M.
AU - Ye, Jieping
AU - Wang, Yalin
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2017/6/15
Y1 - 2017/6/15
N2 - Cortical thickness estimation performed in vivo via magnetic resonance imaging (MRI) is an important technique for the diagnosis and understanding of the progression of Alzheimer's disease (AD). Directly using raw cortical thickness measures as features with Support Vector Machine (SVM) for clinical group classification only yields modest results since brain areas are not equally atrophied during AD progression. Therefore, feature reduction is generally required to retain only the most relevant features for the final classification. In this paper, a spherical sparse coding and dictionary learning method is proposed and it achieves relatively high classification results on publicly available data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) 2 dataset (N = 201) which contains structural MRI data of four clinical groups: cognitive unimpaired (CU), early mild cognitive impairment (EMCI), later MCI (LMCI) and AD. The proposed framework takes the estimated cortical thickness and the spherical parameterization computed by FreeSurfer as inputs and constructs weighted patches in the spherical parameter domain of the cortical surface. Then sparse coding is applied to the resulting surface patch features, followed by max-pooling to extract the final feature sets. Finally, SVM is employed for binary group classifications. The results show the superiority of the proposed method over other cortical morphometry systems and offer a different way to study the early identification and prevention of AD.
AB - Cortical thickness estimation performed in vivo via magnetic resonance imaging (MRI) is an important technique for the diagnosis and understanding of the progression of Alzheimer's disease (AD). Directly using raw cortical thickness measures as features with Support Vector Machine (SVM) for clinical group classification only yields modest results since brain areas are not equally atrophied during AD progression. Therefore, feature reduction is generally required to retain only the most relevant features for the final classification. In this paper, a spherical sparse coding and dictionary learning method is proposed and it achieves relatively high classification results on publicly available data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) 2 dataset (N = 201) which contains structural MRI data of four clinical groups: cognitive unimpaired (CU), early mild cognitive impairment (EMCI), later MCI (LMCI) and AD. The proposed framework takes the estimated cortical thickness and the spherical parameterization computed by FreeSurfer as inputs and constructs weighted patches in the spherical parameter domain of the cortical surface. Then sparse coding is applied to the resulting surface patch features, followed by max-pooling to extract the final feature sets. Finally, SVM is employed for binary group classifications. The results show the superiority of the proposed method over other cortical morphometry systems and offer a different way to study the early identification and prevention of AD.
KW - Alzheimer's Disease
KW - Cortical Thickness
KW - Sparse Coding
KW - Support Vector Machine (SVM)
KW - Weighted Spherical Harmonics
UR - http://www.scopus.com/inward/record.url?scp=85023174597&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85023174597&partnerID=8YFLogxK
U2 - 10.1109/ISBI.2017.7950557
DO - 10.1109/ISBI.2017.7950557
M3 - Conference contribution
AN - SCOPUS:85023174597
T3 - Proceedings - International Symposium on Biomedical Imaging
SP - 446
EP - 450
BT - 2017 IEEE 14th International Symposium on Biomedical Imaging, ISBI 2017
PB - IEEE Computer Society
T2 - 14th IEEE International Symposium on Biomedical Imaging, ISBI 2017
Y2 - 18 April 2017 through 21 April 2017
ER -