Abstract
Neuroimaging data as 18F-FDG PET is widely used to assist the diagnosis of Alzheimer’s disease (AD). Looking for regions with hypoperfusion/ hypometabolism, clinicians may predict or corroborate the diagnosis of the patients. Modern computer aided diagnosis (CAD) systems based on the statistical analysis of whole neuroimages are more accurate than classical systems based on quantifying the uptake of some predefined regions of interests (ROIs). In addition, these new systems allow determining new ROIs and take advantage of the huge amount of information comprised in neuroimaging data. A major branch of modern CAD systems for AD is based on multivariate techniques, which analyse a neuroimage as a whole, considering not only the voxel intensities but also the relations among them. In order to deal with the vast dimensionality of the data, a number of feature extraction methods have been successfully applied. In this work, we propose a CAD system based on the combination of several feature extraction techniques. First, some commonly used feature extraction methods based on the analysis of the variance (as principal component analysis), on the factorization of the data (as non-negative matrix factorization) and on classical magnitudes (as Haralick features) were simultaneously applied to the original data. These feature sets were then combined by means of two different combination approaches: i) using a single classifier and a multiple kernel learning approach and ii) using an ensemble of classifier and selecting the final decision by majority voting. The proposed approach was evaluated using a labelled neuroimaging database along with a cross validation scheme. As conclusion, the proposed CAD system performed better than approaches using only one feature extraction technique. We also provide a fair comparison (using the same database) of the selected feature extraction methods.
Keywords: Alzheimer's disease, Computer aided diagnosis systems, Dimensionality reduction, Machine learning, Support Vector Machine, 18F-FDG PET.
Current Alzheimer Research
Title:Combining Feature Extraction Methods to Assist the Diagnosis of Alzheimer’s Disease
Volume: 13 Issue: 7
Author(s): F. Segovia, J. M. Górriz, J. Ramírez, C. Phillips and for the Alzheimer’s Disease Neuroimaging Initiative
Affiliation:
Keywords: Alzheimer's disease, Computer aided diagnosis systems, Dimensionality reduction, Machine learning, Support Vector Machine, 18F-FDG PET.
Abstract: Neuroimaging data as 18F-FDG PET is widely used to assist the diagnosis of Alzheimer’s disease (AD). Looking for regions with hypoperfusion/ hypometabolism, clinicians may predict or corroborate the diagnosis of the patients. Modern computer aided diagnosis (CAD) systems based on the statistical analysis of whole neuroimages are more accurate than classical systems based on quantifying the uptake of some predefined regions of interests (ROIs). In addition, these new systems allow determining new ROIs and take advantage of the huge amount of information comprised in neuroimaging data. A major branch of modern CAD systems for AD is based on multivariate techniques, which analyse a neuroimage as a whole, considering not only the voxel intensities but also the relations among them. In order to deal with the vast dimensionality of the data, a number of feature extraction methods have been successfully applied. In this work, we propose a CAD system based on the combination of several feature extraction techniques. First, some commonly used feature extraction methods based on the analysis of the variance (as principal component analysis), on the factorization of the data (as non-negative matrix factorization) and on classical magnitudes (as Haralick features) were simultaneously applied to the original data. These feature sets were then combined by means of two different combination approaches: i) using a single classifier and a multiple kernel learning approach and ii) using an ensemble of classifier and selecting the final decision by majority voting. The proposed approach was evaluated using a labelled neuroimaging database along with a cross validation scheme. As conclusion, the proposed CAD system performed better than approaches using only one feature extraction technique. We also provide a fair comparison (using the same database) of the selected feature extraction methods.
Export Options
About this article
Cite this article as:
Segovia F., M. Górriz J., Ramírez J., Phillips C. and for the Alzheimer’s Disease Neuroimaging Initiative , Combining Feature Extraction Methods to Assist the Diagnosis of Alzheimer’s Disease, Current Alzheimer Research 2016; 13 (7) . https://dx.doi.org/10.2174/1567205013666151116141906
DOI https://dx.doi.org/10.2174/1567205013666151116141906 |
Print ISSN 1567-2050 |
Publisher Name Bentham Science Publisher |
Online ISSN 1875-5828 |
Call for Papers in Thematic Issues
New Advances in the Prevention, Diagnosis, Treatment, and Rehabilitation of Alzheimer's Disease
Aims and Scope: Introduction: Alzheimer's disease (AD) poses a significant global health challenge, with an increasing prevalence that demands concerted efforts to advance our understanding and strategies for prevention, diagnosis, treatment, and rehabilitation. This thematic issue aims to bring together cutting-edge research and innovative approaches from multidisciplinary perspectives to address ...read more
Current updates on the Role of Neuroinflammation in Neurodegenerative Disorders
Neuroinflammation is an invariable hallmark of chronic and acute neurodegenerative disorders and has long been considered a potential drug target for Alzheimer?s disease (AD) and dementia. Significant evidence of inflammatory processes as a feature of AD is provided by the presence of inflammatory markers in plasma, CSF and postmortem brain ...read more
Deep Learning for Advancing Alzheimer's Disease Research
Alzheimer's disease (AD) poses a significant global health challenge, with an increasing number of individuals affected yearly. Deep learning, a subfield of artificial intelligence, has shown immense potential in various domains, including healthcare. This thematic issue of Current Alzheimer Research explores the application of deep learning techniques in advancing our ...read more
Diagnostic and therapeutic biomarkers of dementia
Dementia affects 18 million people worldwide. Dementia is a syndrome of symptoms caused by brain disease, usually chronic or progressive, clinically characterized by multiple impairments of higher cortical functions such as memory, thinking, orientation, and learning. In addition, in the course of dementia, cognitive deficits are observed, which often hinder ...read more
- Author Guidelines
- Graphical Abstracts
- Fabricating and Stating False Information
- Research Misconduct
- Post Publication Discussions and Corrections
- Publishing Ethics and Rectitude
- Increase Visibility of Your Article
- Archiving Policies
- Peer Review Workflow
- Order Your Article Before Print
- Promote Your Article
- Manuscript Transfer Facility
- Editorial Policies
- Allegations from Whistleblowers
- Announcements
Related Articles
-
Chemical and Pharmacological Aspects of Heteroaryl-Nitrones
Current Medicinal Chemistry Is Alzheimers Disease a Myth? When is Disease a Disease?
Current Alzheimer Research Cholesterol Homeostasis Failure in the Brain: Implications for Synaptic Dysfunction and Cognitive Decline
Current Medicinal Chemistry Fibrillar β-Amyloid Impairs the Late Phase of Long Term Potentiation
Current Alzheimer Research Vascular Disease and Risk Factors in Chronic Kidney Disease
Vascular Disease Prevention (Discontinued) Impaired Insulin Sensitivity and Secretion in Patients with Alzheimer’s Disease: The Relationship with Other Atherosclerosis Risk Factors
Current Vascular Pharmacology Triggering Receptor Expressed on Myeloid Cells (TREM) Family and the Application of Its Antagonists
Recent Patents on Anti-Infective Drug Discovery Targeting Neurotrophin Receptors in the Central Nervous System
CNS & Neurological Disorders - Drug Targets Acetylcholinesterase: Molecular Modeling with the Whole Toolkit
Current Topics in Medicinal Chemistry The Effects of Antihypertensive Therapy on Haemostatic Parameters
Current Pharmaceutical Design Current Quest in Natural Bioactive Compounds for Alzheimer’s Disease: Multi-Targeted-Designed-Ligand Based Approach with Preclinical and Clinical Based Evidence
Current Drug Targets Approaching Neurological Diseases to Reduce Mobility Limitations in Older Persons
Current Pharmaceutical Design Neuronal Death in Amyotrophic Lateral Sclerosis (ALS): What Can We Learn from Genetics?
CNS & Neurological Disorders - Drug Targets Oxidative/Nitrosative Stress and Immuno-inflammatory Pathways in Depression: Treatment Implications
Current Pharmaceutical Design Diabetic Cardiomyopathy: Clinical and Metabolic Approach
Current Vascular Pharmacology Multi-Target-Directed Ligands and other Therapeutic Strategies in the Search of a Real Solution for Alzheimer’s Disease
Current Neuropharmacology Insights Into the Molecular Aspects of Neuroprotective Bacoside A and Bacopaside I
Current Neuropharmacology Epigenetic Drugs for Multiple Sclerosis
Current Neuropharmacology Multigenic Control of Drug Response and Regulatory Decision-Making in Pharmacogenomics: The Need for an Upper-Bound Estimate of Genetic Contributions
Current Pharmacogenomics Subclinical Thyroid Dysfunction and the Risk of Cardiovascular Disease
Current Pharmaceutical Design