PREDICITVE 1St.Mary’s Group Of Institutions Guntur,India 2K L University,GreenFields,Vaddeswaram,India




Raveendra Reddy Enumula1, J. Amudhavel2

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

1St.Mary’s Group
Of Institutions Guntur,India


2K L University,GreenFields,Vaddeswaram,India     Corresponding Author: [email protected]




Alzheimer’s disease (AD) is a parlous brain disorder which effects the
abnormal functionality of human brain .There has been less research towards the
diagnosis and early detection of this disease in the last decade.. This work
scrutinizes machine learning methods focused at the early identification of AD,
and prediction of progression in mild cognitive impairment. Data are acquired
from the Alzheimer’s disease Neuroimaging Initiative (ADNI). Multi-region
analysis of cross-sectional and longitudinal FDG-PET images from ADNI are Performed.
Information gathered from FDG-PET images secured at a single time point is used
to improve classification results akin with those gathered using data from
research quality MRI.



Keywords: Alzheimer’s disease,
classification, clustering, Image processing, Machine Learning.





Alzheimer’s disease (AD), having a name after the German physician Alois
Alzheimer, is a condition determined by continuous dementia and the sufficient
existence in the brain of peculiar neuro pathological structures. The primitive
symptom is generally memory loss, followed by further functional and cognitive
decline, such that patients become gradually less able to perform even basic
tasks (de Leon, 1999)12. AD is the most usual cause of dementia in the
elderly, with a worldwide ubiquity that is expected to
growth, as the population ages, from the 26.6 million reported in 2006 to over
100 million by 2050 (Brook Meyer et al., 2007)13.


A diagnosis of AD is produce according to consensus criteria such as the
NINCDS-ADRDA Alzheimer’s Criteria (McKhann et al., 1984) 14, which prepare
guidelines for the classification of patients as having definite, probable, or
possible AD. A diagnosis of definite AD requires that neuro
pathological findings be confirmed by a direct analysis of brain tissue
samples, which may be obtained either at autopsy or from a brain biopsy.



These consist of pre-symptomatic diagnosis, differential diagnosis, and
the appraisal and prediction of progression. Research has shown biochemical and
neuroimaging biomarkers to possess diagnostic and prognostic value for AD, and
recently published revisions to the consensus criteria aim to incorporate these
advances (Albert et al., 2011; McKhann et al., 2011; Sperling et al., 2011)15.


A good number of researchers are working to find biological markers
(biomarkers) that indicate the presence of the disease. Adeli et al. (2005a, b)
45 reviewed the studies on computational modeling of AD with related
biomarkers. According to Alzheimer’s disease Neuroimaging Initiative (Mueller
et al., 2005)16, an archetypal AD biomarker should be able to identify
features of the pathophysiologic processes active in AD before symptom onset.
Also, it should be accurate, reliable, valid, and minimally invasive.


Even though a definite biomarker for early detection of AD has yet to be
discovered, significant advances have been made to detect neuropath logical
processes that help identify people at risk of developing dementia. Jack and
Holtzman (2013)17 reviewed several time-dependent models of AD biomarkers
commonly related to the aging. Promising biomarkers currently used in AD
studies are divided into four categories: Biochemistry, genetics,
neurophysiology, and neuroimaging. Biomarkers have to be discovered. Biomarkers
included in the review are simply those that have been proposed so far.
Additional biomarkers will undoubtedly be discovered in the coming years and


Neuroimaging techniques
stipulate a way for clinicians to survey the structural and functional changes
in the brain associated with the enlargement of AD in vivo. Commonly used
modalities involve magnetic resonance imaging (MRI), X-ray computed tomography
(CT), positron emission tomography (PET), single-photon emission computed
tomography (SPECT), and diffusion tensor imaging (DTI). The work occur in this
thesis will focus on PET and MRI, both of which are reported.









Golrokh Mirzaei (et al.) 1 Suggested state-of-the-art review of the
research presented on the diagnosis of AD based on imaging and machine learning
techniques. 1 Authors suggested different segmentation and machine learning
techniques used for the diagnosis of AD are reviewed inclusive of thresholding,
supervised and unsupervised learning, probabilistic techniques, Atlas based
approaches, and fusion of different image modalities.
























             Figure 1: Machine
learning techniques used in the AD studies.


1 Machine learning techniques offer new methods to estimate diagnosis and
clinical outcomes at an individual level. Machine learning
techniques are divided into supervised and unsupervised techniques. Figure 1
shows different supervised and unsupervised techniques used in AD studies.
learning tools have been extensively applied for the identification of
neurologic or neuropsychiatric disorders, especially Alzheimer’s disease (AD)
and its prodrome, mild cognitive impairment (MCI).


Y. Xia (et al.)2 Proposed automated differentiation of AD and FTD from
normal subjects from the perspective of selecting an optimal combination of
Eigen brains to obtain the features that had the best differentiation ability.
2 Report a genetic algorithm-based method to recognize an optimal combination
of eigenvectors so that the resultant features are having ability of successfully splitting
patients with suspected Alzheimer’s disease and front temporal dementia from
normal controls. 2 Compared his approach with standard PCA on a set of 210
clinical cases and improved the performance in separating the dementia types
with an accuracy of 90.0% and a Kappa statistic of 0.849















Figure 2:  Performance comparison between PCA and
proposed feature selection method.



Buscema ( 3 proposed a nonlinear analysis of complex data is a
valid approach in shedding light on the role of NP and NFT in the improvement
of a deteriorate process leading to AD. Our study shows that during life, ANNs
can exactly predict the presence of AD pathology at death. The adaptive systems
had superior accuracy when compared to models of traditional linear statistics.
Both NFTs and senile plaques are primary lesions of AD. Aged plaques include
both diffuse plaques and NPs; usually the majority of senile plaques are the
diffuse type.


( 4 proposed a Prediction
or early-stage diagnosis of Alzheimer’s disease (AD) needs a comprehensive
understanding of the underlying mechanisms of the disease and its progression. Researchers in this area have approached the problem from
multiple directions by attempting to develop (a) neurological (neurobiological
and neurochemical) models, (b) analytical models for anatomical and functional
brain images, (c) analytical feature extraction models for electroencephalograms
(EEGs), (d) classification models for positive identification of AD, and (e)
neural models of memory and memory impairment in AD. This article presents a
state-of-the-art review of research performed on computational modeling of AD
and its markers.


Amezquita-Sanchez ( 5
proposed Mild cognitive impairment (MCI)is a cognitive disorder characterized
by memory impairment, greater than expected by age.
A new methodology is presented to
identify MCI patients during a working memory task using MEG signals. The
met   -hodology consists of four steps:
In step 1, the complete ensemble empirical mode decomposition (CEEMD) is used
to decompose the MEG signal into a set of adaptive sub-bands according to its
contained frequency information.

In step 2, a nonlinear dynamics
measure based on permutation entropy (PE) analysis is employed to analyze the sub
bands and detect feat user to be used for MCI detection. In step 3, an analysis
of variation (ANOVA) is used for feature selection. In step 4, the enhanced
probabilistic neural network (EPNN) classifier is applied to the selected
features to distinguish between MCI and healthy patients. The usefulness and
effectiveness of the proposed methodology are validated using the sensed MEG
data obtained experimentally from 18 MCI and 19 control patients.




Ramon Casanova ( 18 followed here
a method Baseline structural MRI data from intellectual normal and
Alzheimer’s disease (AD) patients from the AD Neuroimaging Initiative database
were used in this study. They assess here the ill-posedness of this
classification problem beyond distinct dimensions and sample sizes and its correlation
to the execution of regularized logistic regression (RLR), linear support
vector machine (SVM) and linear regression classifier (LRC). In adding, these
methods were compared with their principal components space counterparts.

Carlos Cabrall (
19 followed an approach used the voxel intensities (VI) of each brain scan as
the classification features. Orderly to select the subset of features worn by
the classifier, all the trademarks were ranked corresponding to their Mutual
Information (MI) with the class label and the highest ranking attributes were
then stipulated. Let xi ?
?n signify the training
patterns, i = 1,.. P and yi ?
{1, 2, and 3} indicate the corresponding classification. This feature preference
method remains used in both the favourite class ensemble and in the base
classifiers approach.




The early diagnosis of AD and MCI is essential for patient care and
research, and it is widely accepted that preventive measures plays an important
role to delay or alleviate the progression of AD. For the classification task
of different stages of AD progression.The work presented in the image-based
classification of AD and MCI. Classification results for distinguishing between
AD patients and HC may be converging on a glass ceiling since the diagnostic
consensus criteria themselves have an accuracy of around 90%.Multi-region
analyses of cross-sectional and longitudinal FDG-PET images from ADNI are
performed. Information extracted from FDG-PET images acquired at a single time
point is used to achieve classification results comparable with those obtained
using data from research quality MRI.