Analysis of MRI image data for Alzheimer disease detection using deep learning techniques

• Published in: Multimedia tools and applications Vol. 83; no. 6; pp. 17729 - 17752
• Main Authors: Pradhan, Nilanjana, Sagar, Shrdhha, Singh, Ajay Shankar
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.02.2024; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Alzheimer's disease; Cancer; Computer Communication Networks; Computer Science; Data Structures and Information Theory; Datasets; Deep learning; Dementia; Electronic health records; Health care; Heart diseases; Machine learning; Magnetic resonance imaging; Medical imaging; Multimedia Information Systems; Neurological diseases; Nucleotides; Schizophrenia; Special Purpose and Application-Based Systems; Track 2: Medical Applications of Multimedia; Single nucleotide polymorphisms (SNPs); Alzheimer's disease (AD); Cerebrospinal fluid (CSF); Deep learning (DL); Electronic health record (EHR)
• ISSN: 1573-7721; 1380-7501; 1573-7721
• DOI: 10.1007/s11042-023-16256-2

Alzheimer's disease (AD) is the leading cause of dementia globally and one of the most serious future healthcare issue. AD is expected to rise from 27 million to 106 million cases in the next four decades impacting one in every 85 people on the planet. For the existing healthcare systems, the most frequent kind of dementia is a significant source of worry. AD usually refers to Untreated Schizophrenia, a degenerative neurological disorder defined by memory loss and disorientation. AD is the world's third greatest cause of mortality, after only heart disease and cancer. It has surpassed cancer as the most dreaded disease on the planet. AD is catastrophic in the long-term run since it slowly but gradually destroys the body's cells. A variety of efforts have been made to employ structural Magnetic Resonance Imaging (MRI) modalities to differentiate between people with AD and their healthy counterparts. These have also been examined as deep learning algorithms for the categorization of MRI data. It is difficult to find patients with modest cognitive decline who may acquire Alzheimer's. As a result, creating deep learning-based disease detection techniques to assist clinicians in detecting prospective Alzheimer's patients is crucial. The performance comparison of the Imaging, Electronic Health Record (EHR), and Single Nucleotide Polymorphisms (SNP) datasets are evaluated using the metrics Accuracy, Sensitivity, Specificity, and Multi Area. Different mistakes are added under the curves for gradient calculation. The research results are as follows: based on standard datasets the results show that the proposed feature selection algorithms discover a sub-optimal minimal level feature set from a larger input feature set for diagnosing Alzheimer's disease, with higher values for system performance in terms of Accuracy as well as losses against training and Accuracy and losses against validation. These results can demonstrate the model's suitability for the purpose.