J. Biomedical Science and Engineering, 2013, 6, 1090-1098 JBiSE
http://dx.doi.org/10.4236/jbise.2013.611137 Published Online November 2013 (http://www.scirp.org/journal/jbise/)
Computer-aided differential diagnosis system for
Alzheimer’s disease based on machine learning with
functional and morphological image features
in magnetic resonance imaging
Yasuo Yamashita1,2, Hidetaka Arimura3*, Takashi Yoshiura4, Chiaki Tokunaga2, Ohara Tomoyuki5,
Koji Kobayashi2, Yasuhiko Nakamura2, Nobuyoshi Ohya2, Hiroshi Honda4, Fukai Toyofuku3
1Graduate School of Medical Science, Kyushu University, Fukuoka, Japan
2Division of Radiology, Department of Medical Technology, Kyusyu University Hospital, Fukuoka, Japan
3Faculty of Medical Science, Kyushu University, Fukuoka, Japan
4Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
5Department of Neuropsychiatry, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
Received 1 September 2013; revised 8 October 2013; accepted 25 October 2013
Copyright © 2013 Yasuo Yamashita et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
Alzheimer’s disease (AD) is a dementing disorder and
one of the major public health problems in countries
with greater longevity. The cerebral cortical thickness
and cerebral blood flow (CBF), which are considered
as morphological and functional image features, re-
spectively, could be decreased in specific cerebral re-
gions of patients with dementia of Alzheimer type.
Therefore, the aim of this study was to develop a com-
puter-aided classification system for AD patients ba-
sed on machine learning with the morphological and
functional image features derived from a magnetic
resonance (MR) imaging system. The cortical thick-
nesses in ten cerebral regions were derived as mor-
phological features by using gradient vector trajec-
tories in fuzzy membership images. Functional CBF
maps were measured with an arterial spin labeling
technique, and ten regional CBF values were obtain-
ed by registration between the CBF map and Talai-
rach atlas using an affine transformation and a free
form deformation. We applied two systems based on
an arterial neural network (ANN) and a support vec-
tor machine (SVM), which were trained with 4 mor-
phological and 6 functional image features, to 15 AD
patients and 15 clinically normal (CN) subjects for
classification of AD. The area under the receiver ope-
rating characteristic curve (AUC) values for the two
systems based on the ANN and SVM with both image
features were 0.901 and 0.915, respectively. The AUC
values for the ANN- and SVM-based systems with the
morphological features were 0.710 and 0.660, respec-
tively, and those with the functional features were
0.878 and 0.903, respectively. Our preliminary results
suggest that the proposed method may have potential
for assisting radiologists in the differential diagnosis
of AD patients by using morphological and functional
image features.
Keywords: Computer-aided Classification (CAD);
Alzheimer’s Disease; Magnetic Resonance Imaging
(MRI); Arterial Spin Labeling (ASL); Fuzzy
Membership Image; Cortical Thickness; Cerebral Blood
Flow (CBF)
1. INTRODUCTION
Alzheimer’s disease (AD) is the most common cause of
dementia in the majority of developed countries [1-5].
AD is associated with morphological and functional
changes, i.e., the atrophy of gray matter in the cerebral
cortex, and the decrease of cerebral blood flow (CBF) in
specific cerebral regions which can be evaluated with
magnetic resonance imaging (MRI) and nuclear medi-
cine examinations obtained by positron-emission tomo-
graphy (PET) or single-photon emission computed tomo-
graphy (SPECT) [6-13]. However, the examinations by
PET and SPECT are more expensive and invasive than
those using MRI. On the other hand, arterial spin labe-
*Corresponding author.

Y. Yamashita et al. / J. Biomedical Science and Engineering 6 (2013) 1090-1098 1091
ling (ASL) is a cheaper and non-invasive MR imaging
technique for the measurement of CBF without using
contrast medium [14,15]. Yoshiura et al. suggested that
the CBF map images measured by the ASL technique
can be used to assist radiologists in the discrimination of
patients with AD [16].
In recent years, various kinds of computer-aided dia-
gnosis (CAD) methods for AD patients have been deve-
loped [17-21]. However, to the best of our knowledge,
there is no CAD system for the classification of AD pa-
tients using machine learning with morphological and
functional image features obtained by MR imaging alone.
Therefore, our purpose in this study was to develop a
computer-aided differential diagnosis system for AD
patients based on machine learning with morphological
and functional image features obtained by MR imaging
without contrast medium.
2. MATERIALS AND METHODS
2.1. Subjects and MR Data
This study was approved by an institutional review board
of the Kyushu University Hospital. We applied our
proposed method to three-dimensional (3D) T1-weighted
MR images of the whole brain and ASL images obtained
from 30 cases, including 15 patients who were clinically
diagnosed with AD by a neuropsychiatrist at Kyushu
University Hospital (age range: 54 - 89 years; mean age:
77 years; Mini-Mental State Examination (MMSE) score:
11 - 25; mean: 22) and 15 cognitively normal (CN) sub-
jects (age range: 68 - 86 years; mean age: 73 years;
MMSE score: 28 - 30; mean: 29). These data were acqui-
red on a 3.0-T MRI scanner (Intera Achieva 3.0 T Qua-
sar Dual R2.1; PHILIPS Electronics, Best, Netherlands).
T1-weighted sequencing was performed using a mag-
netization prepared rapid gradient echo (MPRAGE) se-
quence (time of repetition (TR): 8.3 ms; time of echo
(TE): 3.8 ms; time of inversion (TI): 240 ms; flip angle:
8 degrees; sensitivity encoding (SENSE) factor: 2; num-
ber of samples averaged (NAS): 1; 240 × 240 × 150
voxels; individual voxel size: 1.0 mm × 1.0 mm × 1.0
mm). ASL was performed using quantitative signal tar-
geting by alternating radiofrequency pulses labeling of
arterial regions (QUASAR), a pulsed ASL technique de-
veloped by Petersen et al. [22]. The QUASAR protocol
consisted of two-dimensional image sequencing (labeling
slab thickness: 150 mm; gap between the labeling and
imaging slabs: 15 mm; SENSE factor: 2.5; TR: 4000 ms;
TE: 22 ms; sampling interval: 300 ms; sampling time
points: 13; 64 × 64 matrix; individualvoxel size: 3.6 mm
× 3.6 mm; 84 dynamics; seven transverse slices of 6.0
mm thickness (gap: 2 mm)). T2-weighted images (TR:
3000 ms; TE: 105 ms; 512 × 512 matrix; seven tran-
sverse slices of 6.0 mm thickness) were obtained at the
same slice level as the ASL sequence.
2.2. Proposed Method
Our proposed method consists of three steps, i.e., the
measurement of the functional and morphological image
features, and the classification of AD patients based on
machine learning. Figure 1 shows the overall scheme for
the calculation of AD patients and CN subjects based on
the functional and morphological image features. The
average CBFs in 16 cerebral cortical regions were deter-
mined as functional image features based on the CBF
map images obtained by the ASL technique. The average
thicknesses in ten cerebral cortical regions were mea-
sured as morphological image features in 3D T1-weigh-
ted whole brain images. In the next step, a combination
of functional and morphological image features for
classification of AD patients was selected based on the
statistical p-values and post studies in 16 average CBFs
and ten cerebral thicknesses. Finally, AD patients and
CN subjects were classified by using a machine learning
technique, i.e., an arterial neural network (ANN) or a
support vector machine (SVM).
2.2.1. Measurement of Functional Image Features
Average CBFs in 16 cerebral cortical regions were de-
termined as functional image features based on the CBF
map image, which was non-linearly aligned with the Ta-
lairach brain atlas by using a registration method with an
affine transformation and a free form deformation (FFD).
The Talairach brain atlas is one of the standard models
labeled for functional human brain mapping, and consists
Functional image features
Creation of a 3D fuzzy
Membership map for a
Cerebral cortical region
Classification of AD patients based on machine learing
Measurement of average
CBF value in cortical
region in 16 lobar regions Measurement of the
cerebral cortical
thicknesses
Registration of CBF map
images and Talairach
brain atlas
2D CBF map and
T2-weighted brain images
Sementation of the brain
Parenchymal region
3D T1-weighted whole
Brain images
Morphological image
features
Figure 1. Overall scheme for the calculation of AD patients
and CN subjects based on the functional and the morphological
image features.

Y. Yamashita et al. / J. Biomedical Science and Engineering 6 (2013) 1090-1098 1092
of 5 levels (hemisphere, lobe, gyrus, tissue and cell level)
[23-25]. The lobe and tissue levels were used for
measuring the average CBF in each lobe of the cerebral
cortical region.
We developed the registration method between the
Talairach brain atlas and a CBF map image of a patient
through the corresponding T2-weighted brain image.
Figure 2 illustrates the registration procedure for mea-
surement of the average CBF values in 16 cortical re-
gions. Our registration method was composed of two
steps. In the first step, an affine transformation was
applied as a global registration. The affine transforma-
tion is given by
11 1213
21 2223
10011
n
n
n
n
X
CCC x
YCCCy
 
 

 
 
 





(1)
where n
x
and n are the coordinates in the moving
image (the CBF map image or the Talairach brain atlas),
and n
y
X
and n are the coordinates in the deformed
image. The affine transformation matrix consisting of
11 to 32 was obtained by using a least squares-
method based on a singular value decomposition so that
the feature points in the moving and reference images
corresponded with each other. For determination of the
affine transformation, the minimum and maximum coor-
dinates of the binary images of the moving images and
T2-weighted images were selected as four sets of cor-
responding feature points.
Y
C C
Fused
image
Deformed
images
Affine
transformation
and FFD
Affine
transformation
Reference
images
CBF map image
T2-weight image
Talairach brain atlas
With 16 cortical regions
Moving
images
Figure 2. Flowchart of registration for extraction the average
CBF values in 16 cortical regions.
In the second step, FFD was employed as a non-linear
local registration [26] to register the Talairach brain atlas
with the T2-weightedimages by determining the trans-
formation function based on B-spline functions, from
which a moving vector


,
x
y
DD in a two-dimensional
image was obtained. The moved coordinates


,
X
Y in
the coordinate system in the T2-weighted image were-
defined by




,
,
x
yy
X
YxDD (2)
where x and y are the original coordinates in the Ta-
lairach brain atlas. Sixty-four sets of corresponding
feature points were determined by using a template-
matching technique between the Talairach atlas and 64
subimages (matrix size: 128 × 128) obtained from the
T2-weighted image. Each feature point was determined
as the coordinates where the centers of the template subi-
mage took the maximum cross-correltion coefficient in
the Talairach atlas.
To approximate the moving distance space ,
x
y, we
formulate an approximation function ,
D
x
y as uniform
bicubic B-spline functions, which were defined by using
a control lattice
D

overlaid on the domain . We
assumed that
Ω

is an


3mn
Ω

3 lattice which
spans the integer grid in the domain . Let ij

be the
value of the control point on lattice , located at



,ij
for 1, 0i1m

 and 1, 01nj. The approxi-
mation function


,yDx in the moving distance space
was defined in terms of these control points by
 

33
00
,
xkl ik ji
kl
Dxy BsBt 


 (3)
where 1ix



, 1jy



,
s
xx


,


01yy t

,0,1,2,3k

, and l. The
functions k and are uniform cubic B-spline basis
functions defined as
0, 1,2,3
Bl
B
 
3
0
Bt 1t
6
 (4)

32
1
364
6
tt
Bt   (5)

32
2
3331
6
ttt
Bt


 (6)

3
36
t
Bt (7)
We chose 16 cerebral cortical regions in the Talairach
brain atlas, i.e., frontal, limbic, occipital, parietal, sub-
lobar, temporal lobes, posterior cingulate gyri and pre-
cuneuses in the left and right brain hemispheres after the
registration, where the average CBFs were measured.
2.2.2. Measurement of Morphological Image Feature s
Our method applied cerebral cortical thicknesses as mor-

Y. Yamashita et al. / J. Biomedical Science and Engineering 6 (2013) 1090-1098 1093
phological image features for the classification of AD
patients. Tokunaga et al. developed an automated me-
thod for measuring the 3D cerebral cortical thicknesses
in AD patients based on 3D fuzzy membership maps
derived from T1-weighted images, which includes the
atrophy in the cortical and white matter regions deter-
mined on each cortical surface voxel by using mem-
bership profiles on trajectories of local gradient vectors
in a fuzzy membership map [21]. For measurement of
the cortical thicknesses in ten cerebral regions, we adop-
ted Tokunaga’s method. This method consisted of
mainly three steps as follows:
1) Segmentation of the brain parenchymal region
based on a brain model matching between a brain mask
and a 3D T1-weighted image;
2) Creation of a fuzzy membership map for the cere-
bral cortical region based on the fuzzy c-means (FCM)
clustering algorithm;
3) Calculation of the cerebral cortical thickness using
localized gradient vector trajectories in fuzzy member-
ship maps.
In order to investigate the regional atrophy at the lobe
level, i.e., frontal, temporal, parietal, occipital lobes and
insula for the left and right brain hemisphere, the cere-
bral cortical thicknesses were separately evaluated in ten
lobar regions. The ten lobar regions were obtained by
registration of the lobar model image to each brain
parenchymal image by using the affine transformation
and FFD. The lobar model image was selected from a
probabilistic reference system for the human brain at the
International Consortium for Brain Mapping (ICBM)
website of the Laboratory of Neuro Imaging (LONI)
[27].
2.2.3. Cl assific ation of AD Patients
We applied two machine learning classifiers, i.e., an
ANN and a SVM, which were trained with the functional
and morphological image features, to 15 AD patients and
15 CN subjects for classification of AD. The in putfunc-
tional features for the classifiers were the average CBF
values in the six regions, i.e., the four lobes (left occipital
lobe, left posterior cingulate gyrus, left and right precu-
nei) where AD-related hypoperfusion was found in the
previous step, and the two regions (right occipital lobe
and right parietal lobe) where the hypoperfusion was
expected based on previous reports [28]. In addition, the
input morphological features were the average values of
the cortical region thicknesses in four regions, i.e., the
left and right temporal lobes, and the left and right insula,
all of which showed statistically significant differences
between AD and CN subjects. All input features were
normalized for the training and testing of the classifiers.
Ten input features for the ANN were normalized from
−0.9 to 0.9, because the hyperbolic tangent (tanh) func-
tion was used as a neuron output function. The ANN
with ten inputs, four hidden layers and one output was
trained based on a Levenberg-Marquardt algorithm, in
which the learning coefficient was empirically set as 0.9,
a convergence criterion was empirically set as 0.0001
and the maximum number of iterations was set as 200.
Regarding the SVM, input features were normalized
from −1.0 to +1.0. We constructed an SVM classifier
with a Gaussian kernel by using the open source software
package SVM light [29], which was empirically set as
3.0 for this study. The regularization parameter C of a
cost function for determination of an optimal hyperplane,
which can efficiently distinguish between AD cases and
CN subjects, was empirically determined as 180. The
maximum number of iterations was set as 100,000.
2.2.4. Evaluation of Our Proposed Classification
System for AD
The performance of our proposed method was evaluated
based on a receiver operating characteristic (ROC) analy-
sis, where the area under the ROC curve (AUC) was
used as a measure of the performance for classification
of AD. The ANN and the SVM were trained and tested
using a leave-one-out-by-case method. The ROCKIT
program was used for creating the ROC curve [30]. The
performances of classification of AD patients based on
an ANN and a SVM were compared with each other. In
addition, the performances using the morphological and/
or functional image features were compared with those
using one of two kinds of image features to investigate
the effect of the image features.
The statistical differences in CBFs and cortical thick-
nesses between AD patients and CN subjects in each
lobe were estimated with the Student paired t test.
3. RESULTS
Figures 3(a) and (b) show therelationship between the
average CBFs and cortical thicknesses in the frontal lobe
and temporal lobe, respectively. The relationships bet-
ween the CBFs and cortical thicknesses of the AD
patients and CN subjects in the frontal lobe and the
temporal lobe were overlapped in their feature spaces.
There were no statistically significant differences be-
tween the two groups in the average CBF or cortical
thicknesses of frontal lobe. On the other hand, there were
statistically significant differences between the AD pa-
tients and the CN subjects in the average CBF of pre-
cuneus (p < 0.05) and cortical thicknesses of the tem-
poral lobe (p < 0.05). Table 1 shows the results of
average CBFs and average thicknesses in cortical regions
of each lobe. The average CBFs and cortical thicknesses
of AD patients were 29.3 ml/100ml/min and 3.15 mm,
respectively. On the other hand, the average CBFs and
cortical thicknesses of CN subjects were 33.1 ml/100

Y. Yamashita et al. / J. Biomedical Science and Engineering 6 (2013) 1090-1098 1094
Table 1. The results of average CBFs and average thicknesses in cortical regions of each lobe.

Y. Yamashita et al. / J. Biomedical Science and Engineering 6 (2013) 1090-1098 1095
0
10
20
30
40
50
012345
Average CBF of frontal lobe
(ml/100ml/min)
Average thickness of cortical
region in frontal lobe (mm)
AD
CN
(a)
0
10
20
30
40
50
012345
Average CBF of precuneus
(ml/100ml/min)
Average thickness of cortical
region in temporal lobe (mm)
AD
CN
(b)
Figure 3. Relationship between the ave-
rage CBFs and cortical thicknesses in the
frontal lobe (a) and temporal lobe (b).
There were no significant differences be-
tween the two groups in either the average
CBF or the cortical thicknesses in the
frontal lobe (a). On the other hand, there
were statistically significant differences
between the AD patients and the CN sub-
jects in the average CBF of the precuneus
(p < 0.05) and cortical thicknesses of the
temporal lobe (p < 0.05) (b).
ml/min and 3.50 mm, respectively. In addition, there
were statistically significant differences between the AD
patients and the CN subjects in the average CBFs of the
left occipital lobe, left posterior cingulate gyrus, left pre-
cuneus, and right precuneus and in the cortical thick-
nesses of the left and right temporal lobe, and left and
right insula (p < 0.05).
Figure 4 shows ROC curves for the overall perfor-
mance of our method in classifying patients with AD and
CN subjects by using the ANN system and the SVM
system. The AUC values for the ANN- and the SVM-
based systems using both image features were 0.901 and
0.915, respectively. The AUC values for the ANN- and
SVM-based systems with the morphological features
AUC=0.878
AUC=0.710
AUC=0.901
(a)
AUC=0.660
AUC=0.915
AUC=0.903
(b)
Figure 4. Receiver-operating characteris-
tic curves for overall performance of our
method in classification of patients with
AD and CN subjects by using the ANN
system (a) or the SVM system (b). The
areas under the curve in both classifier
systems were improved to over 0.9 when
calculated using the average CBFs and
thicknesses. The area under the curve in
the SVM system was particularly impro-
ved, to 0.915, when the average CBFs and
thicknesses were used.
were 0.710 and 0.660, respectively, and those with the
functional features were 0.878 and 0.903, respectively.
4. DISCUSSION
This study showed that the proposed CAD system based
on a combination of the morphological and functional
image features yielded a higher diagnostic performance
for classification of AD compared with those using only
of the two kinds of image features. Although the pro-
posed method misclassified two AD patients and a CN
subject when using only one type of features, the pro-
posed method correctly identified three cases when both
the functional and morphological image features were

Y. Yamashita et al. / J. Biomedical Science and Engineering 6 (2013) 1090-1098 1096
used. Figure 5 shows one (an 81-year-old male with an
MMSE score of 20) of the two AD cases that were mis-
classified using only the cortical thickness features.
However, the proposed method correctly classified this
AD case when a combination of the functional and mor-
phological image features was used, which may have
been due to thelow CBF value, as shown in Figure 5.
Arimura et al. [17] developed a CAD method for AD
with measuring cerebral cortical thicknesses based on
normal vectors in 3D T1-weighted MR image. The AUC
value in their method was 0.909 in a leave-one-out test
method in identification of AD cases among 29 AD cases
(mean age: 70; mean MMSE: 20) and 19 CN cases
(mean age: 62; mean MMSE: 28). Klӧppel et al. [18]
proposed a CAD method by using a linear SVM to
classify the grey matter segment of T1-weighted MR
scans, and tested their method for distinguishing AD
from CN cases. According to their results, 95% (a sen-
sitivity of 95.0% and a specificity of 95.0%) of AD
patients were distinguished in a leave-one-out test among
20 AD cases (mean age: 81; mean MMSE: 17) and 20
CN cases (mean age: 80; mean MMSE: 29). Colliot et al.
[19] reported that their developed method based on
hippocampal volumes in 3D T1-weightedMR images
achieved a classification rate of 84% (a sensitivity of
84%, a specificity of 84%, and a AUC value of 0.913)
between 25 AD patients (mean age: 73; mean MMSE:
24) and 25 controls (mean age: 64; MMSE: no descrip-
tion). Ramirez et al. [20] developed a CAD system for
AD patients based on a baseline principal component
analysis (PCA) system in brain SPECT images. They re-
0
140
(ml/ 100g/mi n)
(a) (b)
0.0
6.0
(mm)
()
()
()
(d)
()
(c) (d) (e)
Figure 5. Original MR images and color-coded maps of cere-
bral cortical thicknesses in patients with Alzheimer’s disease
(age: 81 years; gender: male; mini-mental statement examina-
tion score: 20): (a) an original T2-weighted image; (b) an ori-
ginal CBF map image obtained by the ASL technique; (c) an
original T1-weighted image; (d) a color-coded axial map of
cortical thicknesses; (e) a color-coded volume-rendering map
of cortical thicknesses.
ported a sensitivity of 100%, a specificity of 92.7%, and
an accuracy of 96.9% for 41 AD cases and 56 CN cases
(age and MMSE were not mentioned). On the other hand,
in our results, the AUC values using the SVM-based sys-
tem when individually using morphological features and
functional features were 0.660 and 0.903, respectively. In
comparison between this study and past studies, the AUC
value of the proposed method was lower than conven-
tional methods when using only morphological image
features. However, the AUC value achieved 0.915 when
applying the combination of two image features.
The proposed CAD system for differential diagnosis
of AD has the advantage that it can provide the func-
tional and morphological image features by means of
only an MR examination without contrast medium. If the
differential diagnostic accuracy of AD could be impro-
ved by using our proposed system, then highly accurate
AD diagnosis would be achievable by only an MR
examination without contrast medium, and the exami-
nation burden for patients would be mitigated.
Our proposed method has three limitations. The first
limitation is that the classification results were affected
bysome artifacts on MR imaging. Such artifacts were
particularly prevalent when using the ASL technique,
and included motion artifacts, N/2 ghost artifacts, which
area type of magnetic susceptibility artifacts, and the
artifacts caused by blood flow in the vessels. Figure 6
shows an AD case (a 72-year-old man with an MMSE
score of 23) that was incorrectly classified as a CN sub-
ject due to overestimation of the CBF value by motion
artifact. The second limitation was the number of cases
0
140
(ml / 100g/ m i n)
(a) (b)
0.0
6.0
(mm)
(c) (d) (e)
Figure 6. Original MR images and color-coded maps of cere-
bral cortical thicknesses in patients with Alzheimer’s disease
(age: 72 years; gender: male; mini-mental statement examina-
tion score: 23): (a) an original T2-weighted image; (b) an
original CBF map image obtained by the ASL technique; (c)
an original T1-weighted image; (d) a color-coded axial map
of cortical thicknesses; (e) a color-coded volume-rendering
map of cortical thicknesses.

Y. Yamashita et al. / J. Biomedical Science and Engineering 6 (2013) 1090-1098 1097
used to train the classifier. Both machine learning classi-
fiers, SVM and ANN, were trained with 15 AD cases
and 15 CN subjects in our proposed method. It will be
necessary to collect more data sets in order to improve
the classification accuracy, because the number of
training cases greatly influences the result [31]. The third
limitation is that additional classifiers will need to be
tested, because only the SVM and ANN were evaluated
in this study.
5. CONCLUSION
We have developed a computer-aided classification sys-
tem for AD patients based on a combination of mor-
phological and functional image features obtained by
MR imaging without contrast medium. Our preliminary
results suggest that the proposed method may have
feasibility for the classification of AD patients by using
morphological and functional image features.
6. ACKNOWLEDGEMENTS
The authors are grateful to all members of the Arimura-Laboratory
(http://www.shs.kyushu-u.ac.jp/~arimura) for their valuable comments
and helpful discussion. The authors thank the members of the Division
of Radiology of the Department of Medical Technology at Kyushu
University Hospital. This work was supported by JSPS KAKENHI
Grant Number 21791199.
REFERENCES
[1] World Health Organization (2006) The World Health
Report 2006, Working Together for Health.
http://www.who.int/whr/2006/en/index.html
[2] Chris, H., Vikas, S., Guofan, X. and Sterling C.J. (2011)
The Alzheimer’s disease neuroimaging initiative, predic-
tive markers for AD in a multi-modality framework: An
analysis of MCI progression in the ADNI population.
NeuroImage, 55, 574-589.
http://dx.doi.org/10.1016/j.neuroimage.2010.10.081
[3] Catriona, D.M., David, C., Stephen P.M. and A. Peter P.
(2001) Risk factors for dementia. Advances in Psychiatric
Treatment, 7, 24-31. http://dx.doi.org/10.1192/apt.7.1.24
[4] Yoshitake, T., Kiyohara, Y., Kato, I., Ohmura, T., Iwa-
moto, H., Nakayama, K., Ohmori, S., Nomiyama, K.,
Kawano, H., Ueda, K., Sueishi, K., Tsuneyoshi, M. and
Fujishima, M. (1995) Incidence and risk factors of vas-
cular dementia and Alzheimer’s disease in a defined eld-
erly Japanese population. Neurology, 45, 1161-1168.
http://dx.doi.org/10.1212/WNL.45.6.1161
[5] Claire, M. and Christian, D. (2006) Alzheimer disease:
Progress or profit? Nature Medicine, 12, 780-784.
http://dx.doi.org/10.1038/nm0706-780
[6] Bronge, L., Bogdanovic, N. and Wahlund, L.O. (2002)
Postmortem MRI and histopathology of white matter
changes in Alzheimer brains: A quantitative, comparative
study. Dementia and Geriatric Cognitive Disorders, 13,
205-212. http://dx.doi.org/ 10.1159/000057698
[7] Brun, A. and Englund, E. (1986) A white matter disorder
in dementia of the Alzheimer type: A pathoanatomical
study. Annals of Neurology, 19, 253-262.
http://dx.doi.org/10.1002/ana.410190306
[8] Englund, E. (1998) Neuropathology of white matter
changes in Alzheimer’s disease and vascular dementia.
Dementia and Geriatric Cognitive Disorders, 9, 6-12.
http://dx.doi.org/10.1159/000051183
[9] Agosta, F., Pievani, M., Sala, S., Geroldi, C., Galluzzi, S.,
Frisoni, G.B. and Filippi, M., (2011) White matter dam-
age in Alzheimer disease and its relationship to gray
matter atrophy. Radiology, 258, 853-863.
http://dx.doi.org/10.1148/radiol.10101284/-/DC1
[10] Matsuda, H. (2007) Cerebral blood flow and metabolic
abnormalities in Alzheimer’s disease. Annals of Nuclear
Medicine, 15, 85-92.
http://dx.doi.org/10.1007/BF02988596
[11] Hirata, Y., Matsuda, H. and Nemoto, K. (2005) Voxel-
based morphometry to discriminate early Alzheimer’s
disease from controls. Neuroscience Letter, 382, 269-274.
http://dx.doi.org/10.1016/j.neulet.2005.03.038
[12] Matsuda, H. (2007) Role of neuroimaging in Alzheimer’s
disease, with emphasis on brain perfusion SPECT. Jour-
nal of the Nuclear Medicine, 48, 1289-1300.
http://dx.doi.org/10.2967/jnumed.106.037218
[13] Li, S., Shi, F., Pu, F., Li, X., Jiang, T., Xie, S. and Wang,
Y. (2007) Hippocampal shape analysis of Alzheimer dis-
ease based on machine learning methods. American
Journal of Neuroradiology, 28, 1339-1345.
http://dx.doi.org/10.3174/ajnr.A0620
[14] Petersen, E.T., Zimine, I., Ho, Y.C. and Golay, X. (2006)
Non-invasive measurement of perfusion: A critical re-
view of arterial spin labelling techniques. British Journal
of Radiology, 79, 688-701.
http://dx.doi.org/10.1259/bjr/67705974
[15] Kim, S.G. and Tsekos, N.V. (1997) Perfusion imaging by
a flow-sensitive alternating inversion recovery (FAIR)
technique: Application to functional brain imaging. Ma-
gnetic Resonance in Medicine, 37, 425-35.
http://dx.doi.org/10.1002/mrm.1910370321
[16] Yoshiura, T., Hiwatashi, A., Noguchi, T., Yamashita, K.,
Ohyagi, Y., Monji, A., Nagao, E., Kamano, H., Togao, O.
and Honda, H. (2009) Arterial spin labelling at 3-T MR
imaging for detection of individuals with Alzheimer’s
disease. European Radiology, 19, 2819-2825.
http://dx.doi.org/10.1007/s00330-009-1511-6
[17] Arimura, H., Yoshiura, T., Kumazawa, S., Tanaka, K.,
Koga, H., Mihara, F., Honda, H., Sakai, S., Toyofuku, F.
and Higashida, Y. (2008) Automated method for identi-
fication of patients with Alzheimer’s disease based on
three-dimensional MR images. Academic Radiology, 15,
274-284. http://dx.doi.org/10.1016/j.acra.2007.10.020
[18] Klöppel, S., Stonnington, C.M., Chu, C., Draganski, B.,
Scahill, R.I., Rohrer, J.D., Fox, N.C., Jack, C.R., Ash-
burner, Jr, J. and Frackowiak, R.S.J. (2008) Automatic
classification of MR scans in Alzheimer’s disease. Brain,
131, 681-689. http://dx.doi.org/10.1093/brain/awm319

Y. Yamashita et al. / J. Biomedical Science and Engineering 6 (2013) 1090-1098 1098
[19] Colliot, O., Chételat, G., Chupin, M., Desgranges, B.,
Magnin, B., Benali, H., Dubois, B., Garnero, L., Eustache,
F. and Lehéricy, S. (2008) Discrimination between Alz-
heimer disease, mild cognitive impairment, and normal
aging by using automated segmentation of the hippo-
campus. Radiology, 248, 194-201.
http://dx.doi.org/10.1148/radiol.2481070876
[20] Ramírez, J., Górriz, J.M., Segovia, F., Chaves, R., Salas-
Gonzalez, D., López, M., Álvarez, I. and Padilla, P. (2010)
Computer aided diagnosis system for the Alzheimer’s
disease based on partial least squares and random forest
SPECT image classification. Neuroscience Letters, 472,
99-103. http://dx.doi.org/10.1016/j.neulet.2010.01.056
[21] Tokunaga, C., Arimura, H., Yoshiura, T., Ohara, T., Ya-
mashita, Y., Kobayashi, K., Magome, T., Nakamura, Y.,
Honda, H., Hirata, H., Ohki, M. and Toyofuku, F. (2013)
Automated measurement of three-dimensional cerebral
cortical thickness in Alzheimer’s patients using localized
gradient vector trajectory in fuzzy membership maps.
Journal of Biomedical Science and Engineering, 6, 327-
336. http://dx.doi.org/10.4236/jbise.2013.63A042
[22] Petersen, E.T., Lim, T. and Golay, X. (2006) Model-free
arterial spin labeling quantification approach for perfu-
sion MRI. Magnetic Resonance in Medicine, 55, 219-232.
http://dx.doi.org/10.1002/mrm.20784
[23] Talairach, J. and Tournoux, P. (1988) Co-planar stereo-
taxic atlas of the human brain: 3-Dimensional propor-
tional system: An approach to cerebral imaging. Thieme
Medical Publishers, Inc., New York.
[24] Lancaster, J.L., Woldorff, M.G., Parsons, L.M., Liotti, M.,
Freitas, C.S., Rainey, L., Kochunov, P.V., Nickerson, D.,
Mikiten, S.A. and Fox, P.T. (2000) Automated Talairach
Atlas labels for functional brain mapping. Human Brain
Mapping, 1, 120-131.
http://dx.doi.org/10.1002/1097-0193(200007)10:3<120::
AID-HBM30>3.0.CO;2-8
[25] Lancaster, J.L., Rainey, L.H., Summerlin, J.L., Freitas,
C.S., Fox, P.T., Evans, A.C., Toga, A.W. and Mazziotta
J.C. (1997) Automated labeling of the human brain: A
preliminary report on the development and evaluation of
a forward-transform method. Human Brain Mapping, 5,
238-242.
http://dx.doi.org/10.1002/(SICI)1097-0193(1997)5:4<238
::AID-HBM6>3.0.CO;2-4
[26] Lee, S., Wolberg, G. and Shin, S.Y. (1997) Scattered data
interpolation with multilevel B-splines. IEEE Transac-
tions on Visualization and Computer Graphics, 3, 228-
244. http://dx.doi.org/10.1109/2945.620490
[27] Laboratory of Neuro Imaging (2012) International con-
sortium for brain mapping.
http://www.loni.ucla.edu/ICBM/
[28] Alsop, D.C., Detre, J.A. and Grossman, M. (2000) As-
sessment of cerebral blood flow in Alzheimer’s disease
by spinlabeled magnetic resonance imaging. Annals of
Neurology, 47, 93-100.
http://dx.doi.org/10.1002/1531-8249(200001)47:1<93::AI
D-ANA15>3.0.CO;2-8
[29] Joachims, T. (2008) SVMlight, Cornell University.
http://svmlight.joachims.org/
[30] Metz, C.E., Herman, B.A. and Roe, C.A. (1998) Statisti-
cal comparison of two ROC curve estimates obtained
from partially-paired datasets. Medical Decision Making,
18, 110-121.
http://dx.doi.org/10.1177/0272989X9801800118
[31] Wua, T.K., Huangb, S.C. and Mengc, Y.R. (2008) Eva-
luation of ANN and SVM classifiers as predictors to the
diagnosis of students with learning disabilities. Expert
Systems with Applications, 34, 1846-1856.
http://dx.doi.org/10.1016/j.eswa.2007.02.026