TY - JOUR
T1 - Anatomical Landmark Based Deep Feature Representation for MR Images in Brain Disease Diagnosis
AU - Liu, Mingxia
AU - Zhang, Jun
AU - Nie, Dong
AU - Yap, Pew Thian
AU - Shen, Dinggang
N1 - Funding Information:
Manuscript received August 4, 2017; revised November 22, 2017; accepted January 5, 2018. Date of publication January 10, 2018; date of current version August 31, 2018. This work was supported by the National Institutes of Health under Grants EB006733, EB008374, EB009634, MH100217, AG041721, AG042599, AG010129, and AG030514. (M. Liu and J. Zhang contributed equally to this work.) (Corresponding author: Dinggang Shen.) M. Liu, J. Zhang, D. Nie, and P.-T. Yap are with the Department of Radiology and BRIC, University of North Carolina, Chapel Hill, NC 27599, USA (e-mail: mxliu@med.unc.edu; xdzhangjun@gmail.com; dongnie@cs.unc.edu; ptyap@med.unc.edu).
Publisher Copyright:
© 2013 IEEE.
PY - 2018/9
Y1 - 2018/9
N2 - Most automated techniques for brain disease diagnosis utilize hand-crafted (e.g., voxel-based or region-based) biomarkers from structural magnetic resonance (MR) images as feature representations. However, these hand-crafted features are usually high-dimensional or require regions-of-interest defined by experts. Also, because of possibly heterogeneous property between the hand-crafted features and the subsequent model, existing methods may lead to sub-optimal performances in brain disease diagnosis. In this paper, we propose a landmark-based deep feature learning (LDFL) framework to automatically extract patch-based representation from MRI for automatic diagnosis of Alzheimer's disease. We first identify discriminative anatomical landmarks from MR images in a data-driven manner, and then propose a convolutional neural network for patch-based deep feature learning. We have evaluated the proposed method on subjects from three public datasets, including the Alzheimer's disease neuroimaging initiative (ADNI-1), ADNI-2, and the minimal interval resonance imaging in alzheimer's disease (MIRIAD) dataset. Experimental results of both tasks of brain disease classification and MR image retrieval demonstrate that the proposed LDFL method improves the performance of disease classification and MR image retrieval.
AB - Most automated techniques for brain disease diagnosis utilize hand-crafted (e.g., voxel-based or region-based) biomarkers from structural magnetic resonance (MR) images as feature representations. However, these hand-crafted features are usually high-dimensional or require regions-of-interest defined by experts. Also, because of possibly heterogeneous property between the hand-crafted features and the subsequent model, existing methods may lead to sub-optimal performances in brain disease diagnosis. In this paper, we propose a landmark-based deep feature learning (LDFL) framework to automatically extract patch-based representation from MRI for automatic diagnosis of Alzheimer's disease. We first identify discriminative anatomical landmarks from MR images in a data-driven manner, and then propose a convolutional neural network for patch-based deep feature learning. We have evaluated the proposed method on subjects from three public datasets, including the Alzheimer's disease neuroimaging initiative (ADNI-1), ADNI-2, and the minimal interval resonance imaging in alzheimer's disease (MIRIAD) dataset. Experimental results of both tasks of brain disease classification and MR image retrieval demonstrate that the proposed LDFL method improves the performance of disease classification and MR image retrieval.
KW - Anatomical landmarks
KW - brain disease diagnosis
KW - classification
KW - convolutional neural network
KW - image retrieval
UR - http://www.scopus.com/inward/record.url?scp=85041234710&partnerID=8YFLogxK
U2 - 10.1109/JBHI.2018.2791863
DO - 10.1109/JBHI.2018.2791863
M3 - Article
C2 - 29994175
AN - SCOPUS:85041234710
VL - 22
SP - 1476
EP - 1485
JO - IEEE Journal of Biomedical and Health Informatics
JF - IEEE Journal of Biomedical and Health Informatics
SN - 2168-2194
IS - 5
M1 - 8253440
ER -