Latent feature representation with stacked auto-encoder for AD/MCI diagnosis

Received: 22 July 2013 / Accepted: 14 November 2013 / Published online: 22 December 2013
Springer-Verlag Berlin Heidelberg 2013

abstract

Unlike the previous methods that considered simple lowlevel
features such as gray matter tissue volumes from
MRI, and mean signal intensities from PET, in this paper,
we propose a deep learning-based latent feature representation
with a stacked auto-encoder (SAE)
We believe that
there exist latent non-linear complicated patterns inherent
in the low-level features such as relations among features.

Introduction

Although these researches presented the effectiveness of
their methods in their own experiments on multi-modal
AD/MCI classification, the main limitation of the previous
work is that they considered only simple low-level features
such as cortical thickness and/or gray matter tissue volumes
from MRI mean signal intensities from PET and t-tau, p-tau, and b-amyloid 42 (Ab42) from CSF
In this paper, we assume that there exists hidden or latent highlevel high level information, inherent in those low-level features such
as relations among them, which can be helpful to build a
more robust model for AD/MCI diagnosis and prognosis

The main contributions of our work can be summarized
as follows: (1) To our best knowledge, this is the first work
that considers a deep learning for feature representation in
brain disease diagnosis and prognosis. (2) Unlike the previous
work in the literature, we consider complicated nonlinear
latent feature representation, which can be discovered
from data in self-taught learning. (3) By constructing
an augmented feature vector via a concatenation of the
original low-level features and the SAE-learned latent
feature representation, we can improve diagnostic accuracy
on the public ADNI dataset. (4) By means of pre-training
of SAE in an unsupervised manner with the target-unrelated
samples and then fine-tuning with target-related
samples, the proposed method further enhances the classification
performance.

Materials and image processing

In this work, we use the ADNI dataset publicly available on
the web
http://www.loni.ucla.edu/ADNI

Image processing and feature extraction

The MR images were preprocessed by applying the typical
procedures of anterior commissure (AC)–posterior commissure
(PC) correction, skull-stripping, and cerebellum
removal. Specifically, we used MIPAV software4 for
AC-PC correction, resampled images to 256 9 256 9 256,
and applied N3 algorithm (Sled et al. 1998) to correct
intensity inhomogeneity. An accurate and robust skull
stripping (Wang et al. 2011) was performed, followed by
cerebellum removal. We further manually reviewed the
skull-stripped images to ensure clean and dura removal.
Then, FAST in FSL package5 (Zhang et al. 2001) was used
for structural MR image segmentation into three tissue
types of gray matter (GM), white matter (WM) and cerebrospinal
fluid (CSF). We finally pacellated them into 93
regions of interests (ROIs) by warping Kabani et al.’s
(1998) atlas to each subject’s space via HAMMER (Shen
and Davatzikos 2002), although other advanced registration
methods can also be applied for this process (Friston et al.
1995; Evans and Collins 1997; Rueckert et al. 1999; Shen
et al. 1999;Wu et al. 2006; Xue et al. 2006a, b; Avants et al.
2008; Yang et al. 2008; Tang et al. 2009; Vercauteren et al.
2009; Jia et al. 2010). In this work, we only considered GM
for classification, because of its relatively high relatedness
to AD/MCI compared to WM and CSF (Liu et al. 2012).
4 URL: http://mipav.cit.nih.gov/clickwrap.php.
5 URL: http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/.

Stacked auto-encoder for latent feature representation

In this section, we describe the proposed method for AD/
MCI classification. Figure 1 illustrates a schematic diagram
of the proposed method. Given multi-modal data
along with the class-label and clinical scores, we first
extract low-level features from MRI and FDG-PET as
explained in ‘‘Image processing and feature extraction’’.
We then discover a latent feature representation from the
low-level features in MRI, FDG-PET, and CSF, individually,
by deep learning with SAE.

In deep learning, we
perform two steps sequentially: (1) We first pre-train the
SAE in a greedy layer-wise manner to obtain good initial
parameters. (2) We then fine-tune the deep network to find
the optimal parameters. A sparse learning on the augmented
feature vectors, i.e., a concatenation of the original
low-level features and the SAE-learned features, is applied
to select features that efficiently regress the target values,
e.g., class-label and/or clinical scores.

Sparse auto-encoder

Stacked auto-encoder

Feature selection with sparse representation learning

Multi-kernel SVM learning