Advances in Computerized Analysis in Clinical and Medical Imaging

J Dinesh Peter, Steven Lawrence Fernandes, Carlos Eduardo Thomaz, J Dinesh Peter, Steven Lawrence Fernandes, Carlos Eduardo Thomaz

• Publisher: Chapman and Hall/CRC
• Year: 2019
• Print ISBN: 9781138333291
• eBook ISBN: 9780429820489
• Topic: Computer Science
• Subtopic: Computer Graphics
• Index: Computer Science

Advances in Computerized Analysis in Clinical and Medical Imaging book is devoted for spreading of knowledge through the publication of scholarly research, primarily in the fields of clinical & medical imaging. The types of chapters consented include those that cover the development and implementation of algorithms and strategies based on the use of geometrical, statistical, physical, functional to solve the following types of problems, using medical image datasets: visualization, feature extraction, segmentation, image-guided surgery, representation of pictorial data, statistical shape analysis, computational physiology and telemedicine with medical images.

This book highlights annotations for all the medical and clinical imaging researchers' a fundamental advances of clinical and medical image analysis techniques. This book will be a good source for all the medical imaging and clinical research professionals, outstanding scientists, and educators from all around the world for network of knowledge sharing. This book will comprise high quality disseminations of new ideas, technology focus, research results and discussions on the evolution of Clinical and Medical image analysis techniques for the benefit of both scientific and industrial developments.

Features:

• Research aspects in clinical and medical image processing
• Human Computer Interaction and interface in imaging diagnostics
• Intelligent Imaging Systems for effective analysis using machine learning algorithms
• Clinical and Scientific Evaluation of Imaging Studies
• Computer-aided disease detection and diagnosis
• Clinical evaluations of new technologies
• Mobility and assistive devices for challenged and elderly people

This book serves as a reference book for researchers and doctoral students in the clinical and medical imaging domain including radiologists. Industries that manufacture imaging modality systems and develop optical systems would be especially interested in the challenges and solutions provided in the book. Professionals and practitioners in the medical and clinical imaging may be benefited directly from authors' experiences.

1

A New Biomarker for Alzheimer’s Based on the Hippocampus Image Through the Evaluation of the Surface Charge Distribution

Aldo A. Belardi, Fernandho de O. Freitas, Rodrigo P. Bechelli, and Rodrigo G. G. Piva

Centro Universitário da FEI, Department of Electrical Engineering, Av. Humberto de Alencar Castelo Branco, São Bernardo do Campo, São Paulo, Brazil

1.1 Introduction

Alzheimer’s disease (AD) is the most common type of dementia among the elderly population, with no cure and epidemiological trend projections over the next thirty-five years worldwide.¹

The bibliography shows mild cognitive impairment (MCI)—an AD predecessor phase which a subject undergoes when the initial abnormal functional and structural transformations occur in the brain, given the AD degenerates nerve cells, reducing gray matter volume.^2,3 Some medical imaging and clinical exams can detect these transformations, such as magnetic resonance imaging (MRI) and positron emission tomography (PET). Scientists have struggled to find out that AD and MCI biological markers can help in early disease prediction and adequate treatments. Currently, only a single marker is available that can predict AD.⁴ Instead, several groups of biomarkers are used to achieve high sensitivity and specificity for classification among AD, MCI, and normal groups cognition.⁴

Hippocampus volume is the widely used AD biomarker because this is the region of the brain first affected by the disease. There are researches that deeply investigate the hippocampus, aiming to detect transformations due to AD.⁵

In this chapter, we present a novel approach to identify morphological changes in the hippocampus in microstructural level. Based on the concept that electrostatic charges exert influences on each other, we analyze the hippocampus surface charge density distribution (SCDD). The objective is verified if SCDD can provide information about the local and global morphological transformations occurring in the hippocampus. For this purpose, we developed a model to calculate the SCDD and embedded a tool where a user can calculate the hippocampal masks SCDD from images.^6–8

1.2 Earlier Detection of Alzheimer’s Disease

The first known publication of Alzheimer’s disease was 100 years ago, but major contributions occurred in the last thirty years, when it started to be treated as lethal and common case of dementia.⁹ In 1984, the National Institute of Neurological Disorders and Stroke and the Alzheimer’s Association developed the clinical criterion to diagnose Alzheimer’s in patients already diagnosed with dementia, according to the Diagnostic and Statistical Manual of Mental Disorders.¹⁰ This method, which we call the “old criteria,” was based mainly on neuropsychology, with tests applications to evaluate whether physical and mental disturbances were perceived and associated with characteristic risk factors of AD. Complementary tools were also applied to the old criteria as electroencephalogram (EEG) and computed tomography to detect abnormal structural alterations and brain activities decelerations.¹¹

Evaluation by the old criterion did not allow the definitive conclusion of AD with the patient still alive. Instead, definitive diagnosis would only be possible after autopsy, evaluating the presence of amyloid plaques and neurofibrillary brains.

In the 1990s, scientists searched for effective ways to diagnose AD. In the biochemistry field, studies focused in obtaining biomarkers of amyloid by extracting the spinal brain fluid.^12–14

In the neuroimaging field, studies search for morphological alterations of the brain in MR images correlated with AD, more specifically gray matter (GM). With indication that AD affects several regions of the brain in different ways according to the analyzed group, medial temporal lobe which includes the hippocampus is the region mostly affected with reductions in large vessel.^15,16 MRI scans contributed to the verification of neurofibrillary bundles and GM reduction due to the death of brain cells.

In the late 1990s, MCI was conceptualized as intermediate stage between normal and patient with AD.¹⁷ An individual with MCI may develop AD as well as other types of dementia, stabilized in MCI, or revert to normal control (NC) status. Usually patients with MCI seek specialists because it is at this stage that memory begins to be compromised. It was understood that the study of MCI would be of great importance and one of the main proposals would b...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Preface
• About the Editors
• Contributors
• 1. A New Biomarker for Alzheimer’s Based on the Hippocampus Image Through the Evaluation of the Surface Charge Distribution
• 2. Independent Vector Analysis of Non-Negative Image Mixture Model for Clinical Image Separation
• 3. Rationalizing of Morphological Renal Parameters and eGFR for Chronic Kidney Disease Detection
• 4. Human Computer Interface for Neurodegenerative Patients Using Machine Learning Algorithms
• 5. Smart Mobility System for Physically Challenged People
• 6. DHS The Cognitive Companion for Assisted Living of the Elderly
• 7. Raspberry Pi Based Cancer Cell Detection Using Segmentation Algorithm
• 8. An AAC Communication Device for Patients with Total Paralysis
• 9. Case Studies on Medical Diagnosis Using Soft Computing Techniques
• 10. Alzheimer’s Disease Classification Using Machine Learning Algorithms
• 11. Fetal Standard Plane Detection in Freehand Ultrasound Using Multi Layered Extreme Learning Machine
• 12. Earlier Prediction of Cardiovascular Disease Using IoT and Deep Learning Approaches
• 13. Analysis of Heart Disease Prediction Using Various Machine Learning Techniques
• 14. Computer-Aided Detection of Breast Cancer on Mammograms Extreme Learning Machine Neural Network Approach
• 15. Deep Learning Segmentation Techniques for Checking the Anomalies of White Matter Hyperintensities in Alzheimer’s Patients
• 16. Investigations on Stabilization and Compression of Medical Videos
• 17. An Automated Hybrid Methodology Using Firefly Based Fuzzy Clustering for Demarcation of Tissue and Tumor Region in Magnetic Resonance Brain Images
• 18. A Risk Assessment Model for Alzheimer’s Disease Using Fuzzy Cognitive Map
• 19. Comparative Analysis of Texture Patterns for the Detection of Breast Cancer Using Mammogram Images
• 20. Analysis of Various Color Models for Endoscopic Images
• 21. Adaptive Fractal Image Coding Using Differential Scheme for Compressing Medical Images
• Index