Data Mining and Knowledge Engineering

Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Effective Features Extracting Approach Using MFCC for Automated Diagnosis of Alzheimer's Disease


Affiliations
1 Department of Computer Science and Engineering, University of Menoufiya, Egypt
2 Department of Electronics and Electrical Communications, University of Menoufiya, Egypt
3 University of Menoufiya, Egypt
     

   Subscribe/Renew Journal


This paper proposes a Computer Aided Diagnosis (CAD) system for extracting the most significant features of Alzheimer's disease (AD) using the Mel-Scale Frequency Cepstral Coefficients (MFCC). The features extracted from AD MRI images are effective features used for the diagnosis of AD. To test the proposed model, the database used Open Access Series of Imaging Studies (OASIS) database for Magnetic Resonance Imaging (MRI). First, the images are transformed from 3-D to 1-D, then it will be processed using the approach given in for feature reduction. In addition, the MFCC algorithm will be used to extract the effective features, then applying a Linear Support Vector Machine (SVM) classifier for classification step. This paper presents the comparison study between the approach given in, the approach using MFCC only, and using proposed approach. The obtained results provides excellent results using the proposed algorithm with high accuracy using small number of significant extracted features.

Keywords

Diagnosis, Alzheimer's Disease, Computer Aided Diagnosis, Feature Extraction, Mel-Scale Frequency Cepstral Coefficients, Feature Reduction, and Support Vector Machine.
User
Subscription Login to verify subscription
Notifications
Font Size

Abstract Views: 187

PDF Views: 2




  • Effective Features Extracting Approach Using MFCC for Automated Diagnosis of Alzheimer's Disease

Abstract Views: 187  |  PDF Views: 2

Authors

Mohamed M. Dessouky
Department of Computer Science and Engineering, University of Menoufiya, Egypt
Mohamed A. Elrashidy
Department of Computer Science and Engineering, University of Menoufiya, Egypt
Taha E. Taha
Department of Electronics and Electrical Communications, University of Menoufiya, Egypt
Hatem M. Abdelkader
University of Menoufiya, Egypt

Abstract


This paper proposes a Computer Aided Diagnosis (CAD) system for extracting the most significant features of Alzheimer's disease (AD) using the Mel-Scale Frequency Cepstral Coefficients (MFCC). The features extracted from AD MRI images are effective features used for the diagnosis of AD. To test the proposed model, the database used Open Access Series of Imaging Studies (OASIS) database for Magnetic Resonance Imaging (MRI). First, the images are transformed from 3-D to 1-D, then it will be processed using the approach given in for feature reduction. In addition, the MFCC algorithm will be used to extract the effective features, then applying a Linear Support Vector Machine (SVM) classifier for classification step. This paper presents the comparison study between the approach given in, the approach using MFCC only, and using proposed approach. The obtained results provides excellent results using the proposed algorithm with high accuracy using small number of significant extracted features.

Keywords


Diagnosis, Alzheimer's Disease, Computer Aided Diagnosis, Feature Extraction, Mel-Scale Frequency Cepstral Coefficients, Feature Reduction, and Support Vector Machine.