Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


Presented by: Nada Mohamed

No description

Nada El Rakaiby

on 6 July 2014

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Presented by: Nada Mohamed

Achievements in project1 cont.
1- Texture measures
Standard Deviation
Recognition of Alzheimer's Disease using brain MRI
Recognition of Alzheimer's Disease
The research’s main target is the early recognition of Alzheimer’s disease automatically, without the need for a clinical expert.
The early recognition would be helpful in keeping the disease at a certain stage to avoid deterioration of the case that could result in complete brain damage,

Achievements in project1
Analysis started with 5 AD vs. 5 Normal MRI slices from axial (horizontal selection) orientation.

Achievements in project1 cont.
Features extraction techniques were used to describe brain images such as:

1- Texture measures( mean, standard deviation).
Edge detection then area calculations.
These used techniques were fruitless in recognizing the normal brain from the AD infected brain because of the variance in the results.

Alzheimer’s disease is a brain disease that causes a slow decline in memory, thinking and reasoning skills. It represents a major public health problem.
Structural imaging studies such as Magnetic Resonance Imaging (MRI) have shown that the brains of people with Alzheimer's shrink significantly as the disease progresses.
Research has also shown that shrinkage in specific brain regions such as the hippocampus (horseshoe shaped structure located one in the left brain hemisphere and another in the right brain hemisphere involved in memory forming) may be an early sign of Alzheimer's.  .

Achievements in project1
Presented by:
Belal Mohamed
Nada El Rakaiby
Ahmed Amr
Marwan El Khateeb
Supervised by:

Dr Rania Kadry

Standard Deviation
Achievements in project1 cont.
2- Edge detection and area calculation:
Edge Detection
Achievements in project1
A sample of size 10x10 was cropped from the DCT image which represents all the low spatial frequency values. Zigzag scan was applied to generate a vector of 55 elements.
The purpose of the zigzag scan was to group the low frequency coefficients of the 10x10 matrix, the (0,0) element (top-left) has been set to zero because it could adversely affect classification of structural changes in the images.
Zigzag scan on the 10x10 matrix
Achievements in project1 cont.
Euclidean distance measurement was applied on the vector generated from the zigzag scan as a matching stage to compare between the summarized data.
Since we have used 5 normal images and 5 AD images, we used a normal image as a template and compared it’s vector with all the other 9 images
We have found that the Euclidean distance between normal MR image and another normal MR image is small compared to the Euclidean distance between normal MR image and AD MR Image.
Future Work
Area Calculation
after applying edge detection on 10 images
Low Spatial Frequency components were used by applying the DCT, where large low frequency components indicate the presence of larger structures in the image (important details), while the high frequencies are related to the fine details and noise content of the image (could be discarded).
AD before DCT
Normal before DCT
Normal after DCT
AD after DCT
Increase database of MR images to have more accuracy in results.
Upgrade recognition techniques to not only recognize AD infected brain from normal brain but also detect specifically the stage of AD automatically.
Use other different techniques in recognition and compare between the results for better output.
Try to detect AD not only with AXIAL orientation but also with CORNAL orientation.

MRI scan
Achievements in project1 cont.
Euclidean Distance was then used, which is a similarity measurement that is used to decide how close a vector is to another.
It is evaluated by calculating the distance between two points p and q using the following pythagorian formula :
Full transcript