Temporal Change Analysis Based Recommender System for Alzheimer Disease Classification
Santi Swarup Basa1, Debashis Pradhan2, Lipsa Das3, Abhaya Kumar Panda4, Santosh Kumar Swain5
1Santi Swarup Basa*, Department of Computer Science, North Orissa University, Baripada, Odisha. India.
2Debashis Pradhan, Department of Computer Science, North Orissa University, Baripada, Odisha. India.
3Lipsa Das, Department of Computer Science, North Orissa University, Baripada, Odisha. India.
4Abhaya Kumar Panda, Department of Computer Science Engineering, KIIT Polytechnic, KIIT Deemed to be University, Bhubaneswar, Odisha, India.
5Santosh Kumar Swain, Department of Computer Science Engineering, KIIT Deemed to be University, Bhubaneswar Odisha. India.
Manuscript received on January 12, 2020. | Revised Manuscript received on January 22, 2020. | Manuscript published on February 10, 2020. | PP: 480-488 | Volume-9 Issue-4, February 2020. | Retrieval Number: D1202029420/2020©BEIESP | DOI: 10.35940/ijitee.D1202.029420
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The development of recommender systems gathered momentum due to its relevance and application in providing a personalized recommendation on a product or a service for customer relations management. It has proliferated into medicine and its allied domains for the recommendations on disease prediction/detection, medicine, treatment, and other medical services. This chapter describes a new composite and comprehensive recommender system named Temporal Change Analysis based Recommender System for Alzheimer Disease Classification (TCA-RS-AD) using a deep learning model. Its performance is evaluated on the dataset with T1-weighted MRI clinical temporal data of OASIS and the results were recorded in terms of Precision, Recall, F1-Score and Accuracy, Hamming Loss, Cohens Kappa Coefficient, and Matthews Correlation Coefficient. The improved accuracy of this recommendation model endorses its suitability for its application in the classification of AD.
Keywords: Deep Learning Models, Confusion Matrix, Matthews Correlation Coefficient, Hamming Loss, Cohens Kappa, OASIS dataset.
Scope of the Article: Deep Learning