1. commonly happens in patients’ late stage life (Burns

1.    
Introduction

1.1.  
 Alzheimer’s Disease

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


order now

Alzheimer’s
disease is a brain disorder that progresses over time.  It most commonly happens in patients’ late
stage life (Burns
& Iliffe, 2009). 
It involves the death of nerve cells and thus leads to the reduction of
brain volume (Fig.1), which impact
brain functions.  The most noticeable early stage symptom
includes short-term memory loss.  As the
disease progress, patients may show disturbances in language, orientation,
reasoning and perception (Weiner et al.,
2010).  At the same time,
both the annual incidence of Alzheimer’s disease and the cost of caring or
Alzheimer’s disease are on the rise.  It
is believed that the number of Alzheimer’s disease patients will be doubled by
2050.  The estimated amount of money Americans
spent on Alzheimer’s disease treatment and caring is $259 billion USD in 2017. (Alzheimer’s
Association, 2015)

 

 

 

 

 

 

 

 

 

Fig.1

The figure shows a Alzheimer’s disease brain comparing
to normal brain.  The shrinkage of whole
brain size, particularly Hippocampus as well as the shriveled cortex can be
seen.

 

 

1.2.  
 Traditional
Alzheimer’s Disease Diagnosis

Traditional Alzheimer’s disease is diagnosed clinically
based on the person’s description, medical records and behaviors (Mendez, 2006). 
Doctors cannot guarantee an acceptable accuracy for diagnosis without
close examine of the brain tissue under a microscope, which can only be done
after the patient’s death.

 

1.3.   Machine Learning in Alzheimer’s
Disease Diagnosis

Machine
learning enables computers to learn without specifically programmed
instructions and guidelines (Samuel, 2000).  In the field of Alzheimer’s disease
diagnosis, the aim is to increase data processing ability and decrease the specialty
domain expert knowledge needed to detect brain abnormality at a high accuracy (Plis et al., 2014).  Many studies were done to develop such
models.

 

1.4.   Prior Research

Features
like Brain volume, hippocampus shape and cortex character can be used to diagnose
Alzheimer’s disease utilizing several types of brain images.  Structural MRI and function MRI are among the
most popular.  

Prior
researchers started with segmenting brains to different part, extracting
voxel-wise features.  These features are
represented by vectors of multi-scalar quantifications.  Fan successfully segmented the brain to grey
matter, white matter and computed their corresponding voxel-wise densities (Fan, Shen, Gur, Gur, & Davatzikos, 2007).  These values 
were later on converted to vector form that can be used for
classification.  Further on top of their
work, Lerch et al computed cortical thickness by measuring the gap between
selected points at grey matter and white matter (Lerch et al., 2008). 

Convolutional
Neural Networks were also used for the purpose of Alzheimer’s disease diagnosis
in mainly two directions: analyzing medical record and analyzing
bio-images.  Aston and Gunn first proposed
Convolutional Neural Network in 2D dimension to extract unique features from
MRI slices.  And later, Payan employed a
3D Convolutional Neural Network for this purpose (Payan & Montana, 2015).  Recent work by Hosseini-Asl implemented a 3D
Auto-encoder at the lower layer.  It is
pre-trained by a set of dementia data and the upper fully connected layers will
be used for fine tuning by specific data domain (Hosseini-Asl et al., 2016). 

x

Hi!
I'm Clifton!

Would you like to get a custom essay? How about receiving a customized one?

Check it out