Classification of Alzheimer’s disease from Brain MRI Using Transfer Learning From CNN
Author(s):
ROSHINI S P, Dr. T. Thenmozhi
Keywords:
CNN, MRI, generic feature, PCA, TSNE, Classifier
Abstract
For picture categorization and object recognition, various Convolutional Neural Network (CNN) architectures have been developed. It is a challenging effort for CNN to cope with hundreds of MRI Image slices, each of almost similar nature in a single patient, for the image-based categorization. As a result, employing 2D CNN architecture to categorise a number of patients as AD, MCI, or NC based on 3D MRI becomes a hazy technique. As a result, we have simplified the concept of classifying patients based on 3D MRI while acknowledging the 2D features produced by the CNN framework in order to overcome this issue. We outline our approach for extracting 2D characteristics from an MRI and transforming them so that they can be used in a machine learning algorithm for classification. Our study's findings are displayed of classifying patients into 3 groups. We used pretrained Alexnet CNN or scratched trained CNN as a general feature extractor for 2D images, whose dimensions were decreased using PCA+TSNE. Finally, we classified the images using a straightforward machine learning technique like KNN. Even while the outcome is not particularly stunning, it clearly demonstrates that it can be superior to CNN softmax classification trained from scratch using probability scores. The created feature is easily manipulable and can be enhanced for greater specificity, sensitivity, and accuracy.
Article Details
Unique Paper ID: 159908

Publication Volume & Issue: Volume 9, Issue 12

Page(s): 871 - 875
Article Preview & Download


Share This Article

Join our RMS

Conference Alert

NCSEM 2024

National Conference on Sustainable Engineering and Management - 2024

Last Date: 15th March 2024

Call For Paper

Volume 10 Issue 10

Last Date for paper submitting for March Issue is 25 June 2024

About Us

IJIRT.org enables door in research by providing high quality research articles in open access market.

Send us any query related to your research on editor@ijirt.org

Social Media

Google Verified Reviews