Handbook of Decision Support Systems for Neurological Disorders

Description

Handbook of Decision Support Systems for Neurological Disorders provides readers with complete coverage of advanced computer-aided diagnosis systems for neurological disorders. While computer-aided decision support systems for different medical imaging modalities are available, this is the first book to solely concentrate on decision support systems for neurological disorders. Due to the increase in the prevalence of diseases such as Alzheimer, Parkinson’s and Dementia, this book will have significant importance in the medical field. Topics discussed include recent computational approaches, different types of neurological disorders, deep convolution neural networks, generative adversarial networks, auto encoders, recurrent neural networks, and modified/hybrid artificial neural networks. Handbook of Decision Support Systems for Neurological Disorders Copyright Contributors Preface 1. A review of deep learning-based disease detection in Alzheimer's patients 1.1 Introduction 1.1.1 Alzheimer's disease 1.1.1.1 Stages of Alzheimer's 1.1.1.2 Cause of Alzheimer's 1.1.1.3 Signs and symptoms 1.1.1.4 Anatomical changes in the brain 1.1.2 Deep learning 1.2 Literature review 1.2.1 Neuroimaging techniques used in Alzheimer's detection 1.2.1.1 Magnetic resonance imaging 1.2.1.2 Computer tomography 1.2.1.3 Positron emission tomography 1.2.2 Data resources 1.2.2.1 Alzheimer's disease Neuroimaging Initiative 1.2.2.2 Open Access Series of Imaging Studies 1.2.2.3 Minimal Interval Resonance Imaging in Alzheimer's Disease 1.2.3 Tools available for processing neuroimaging data 1.2.3.1 3D Slicer 1.2.3.2 BrainVoyager 1.2.3.3 CONN 1.2.3.4 FreeSurfer 1.2.3.5 SPM 1.2.4 Issues and challenges with medical imaging 1.2.5 Deep learning for imagery data 1.2.5.1 Layers of the CNN Convolution layer Nonlinearity layer Pooling layer Regularization 1.3 Methods of Alzheimer's detection using neuroimaging data 1.3.1 Feature-based methods 1.3.1.1 Voxel-level features 1.3.1.2 Vertex-level features 1.3.1.3 Hippocampus based 1.3.1.4 Miscellaneous 1.3.2 Deep learning-based methods 1.3.2.1 2D CNN-based methods 1.3.2.2 3D CNN-based methods 1.4 Comparison of detection methods 1.5 Conclusion References 2. Brain tissue segmentation to detect schizophrenia in gray matter using MR images 2.1 Introduction 2.2 Data collection 2.3 Methods 2.3.1 Image enhancement and skull stripping 2.3.1.1 Image enhancement 2.3.1.2 Skull stripping 2.3.2 Heuristic-based thresholding 2.3.3 MRF-MAP estimation 2.3.3.1 Markov random field segmentation 2.3.3.2 Maximum a posteriori 2.3.4 Feature extraction 2.3.5 Classifiers 2.3.5.1 Random forest classifier 2.3.5.2 Random tree classifier 2.4 Results and discussion 2.5 Conclusion References 3. Detection of small tumors of the brain using medical imaging 3.1 Introduction 3.2 Types of medical images used in the application 3.3 Theoretical background of the application 3.4 Description of the application 3.4.1 The architecture of the software application 3.4.2 Image segmentation 3.4.3 The cancer detection algorithm 3.4.4 Scenario and functionalities 3.5 Testing, installing, and using the application 3.6 Conclusions References 4. Fuzzy logic-based hybrid knowledge systems for the detection and diagnosis of childhood autism 4.1 Introduction 4.1.1 Autism spectrum disorder 4.1.2 Factors associated with ASD 4.1.3 Traditional approaches to detection: DSM-V 4.2 The advent of machine learning: a new horizon for autism 4.3 Fuzzy-based systems for autism 4.3.1 Fuzzy logic systems 4.3.2 Neuro-fuzzy systems 4.3.3 Fuzzy cognitive maps 4.4 Moving ahead: virtual reality for autism 4.5 Current scenario and scope of improvement 4.6 Discussion and conclusion List of abbreviations Acknowledgments References 5. Artificial intelligence for risk prediction of Alzheimer's disease: a new promise for community health screening in the old ... 5.1 Introduction 5.2 Etiology and risk factors 5.3 Screening and early detection 5.4 Current methods of early detection 5.5 The rise of AI—a new promise for community health screening 5.5.1 Types of AI methods applicable for AD prediction 5.6 Common algorithms used in ML for dementia and AD detection 5.7 Artificial intelligence methodologies for screening AD: concept examples 5.7.1 Utilization of ML for AD detection 5.7.2 Utilization of DL for AD detection 5.7.3 Utilization of an ANN for AD detection 5.8 Promises and challenges of AI applications for predicting AD 5.9 Conclusions and future direction Acknowledgment References 6. Cost-effective assistive device for motor neuron disease 6.1 Introduction 6.2 Motor neuron diseases 6.2.1 Motor neuron types 6.2.2 Causes of motor neuron diseases 6.2.3 Motor neuron disease types 6.2.3.1 Amyotrophic lateral sclerosis 6.2.3.2 Progressive bulbar palsy 6.2.3.3 Pseudobulbar palsy 6.2.3.4 Primary lateral sclerosis 6.2.3.5 Progressive muscular atrophy 6.2.3.6 Spinal muscular atrophy 6.2.3.7 Congenital SMA with arthrogryposis 6.2.3.8 Kennedy's disease 6.3 System configuration 6.3.1 Hardware required 6.3.1.1 Arduino board 6.3.1.2 Accelerometer 6.3.1.3 RF transmitter and receiver module 6.3.2 Software 6.3.3 System description 6.3.3.1 Sensor placement design 6.3.3.2 Wireless communication module 6.3.3.3 Onscreen scanning keyboard design 6.3.3.4 Text-to-speech conversion 6.4 Experimental results 6.5 Conclusion References 7. EEG signal-based human emotion detection using an artificial neural network 7.1 Introduction 7.1.1 Impact of human/computer interaction with emotional intelligence 7.1.2 Affective computing—state-of-the-art 7.1.3 Mathematical modeling effect of expressions 7.1.4 Affect expressions 7.2 EEG signal data acquisition 7.2.1 EEG recording through electrodes 7.2.2 EEG characteristics 7.2.3 EEG emotion measurement 7.2.4 EEG spectral distribution 7.3 Statistical features extracted from EEG 7.3.1 EEG data preprocessing 7.3.2 Feature extraction 7.3.3 Classification 7.4 Various ANN methods to classify EEG data 7.4.1 Artificial neural network 7.4.2 Multilayer perceptron 7.4.3 ANN-based classifier 7.5 Classification of EEG-based emotion using ANN 7.6 Experimental analysis 7.7 Conclusion References Further reading 8. Multiview decision tree-based segmentation of tumors in MR brain medical images 8.1 Introduction 8.2 Multiview decision tree-based segmentation 8.3 Results and discussion 8.4 Conclusion Acknowledgment References 9. Multiclass SVM coupled with optimization techniques for segmentation and classification of medical images 9.1 Introduction 9.2 Classification of support vector machine 9.2.1 Mathematical formulation of a single-class SVM 9.2.2 Mathematical formulation of a two-class SVM 9.2.3 Multiclass SVM 9.3 Parameter tuning of multiclass SVM using optimization algorithms 9.3.1 Simulated annealing-SVM approach 9.3.2 Genetic algorithm-SVM approach 9.3.3 Crow optimization-SVM approach 9.4 Results and discussion 9.5 Conclusion Acknowledgments References 10. Brain tissues segmentation in magnetic resonance imaging for the diagnosis of brain disorders using a convolutional neural ... 10.1 Introduction 10.2 Materials and methods 10.2.1 MR brain images 10.2.2 Preprocessing 10.2.2.1 Denoising using an NLM filter 10.2.2.2 Skull stripping using a CNN 10.2.3 Brain tissue segmentation 10.2.3.1 CNN architecture 10.2.3.2 Performance analysis 10.3 Results and discussion 10.3.1 Denoising result 10.3.2 Skull-stripping result 10.3.3 Segmentation result 10.4 Conclusion and future work References 11. Fine motor skills and cognitive development using virtual reality-based games in children 11.1 Introduction 11.2 Neurological disorders in children and the role of VR games in children's rehabilitation 11.3 Leap Motion Controller and Unity3D 11.4 Game design and working 11.4.1 Game 1: roller ball 11.4.1.1 Roller ball—level 1 11.4.1.2 Roller ball—level 2 11.4.1.3 Roller ball—level 3 11.4.1.4 Outcomes 11.4.2 Game 2: color sorter 11.4.2.1 Outcome 11.4.3 Game 3: shape sorter 11.4.3.1 Outcome 11.4.4 Game 4: stacking 11.4.4.1 Outcome 11.4.5 Game 5: identifying alphabets and numerals 11.4.5.1 Outcome 11.5 Experimental results and discussion 11.6 Future work 11.7 Conclusions Acknowledgments References 12. A CAD software application as a decision support system for ischemic stroke detection in the posterior fossa 12.1 Introduction 12.2 Computer-aided diagnosis system implementation 12.3 Proposed CAD system for ischemic stroke detection 12.3.1 System functionality 12.3.2 System architecture 12.3.3 The developed graphical user interface 12.4 Experimental results and discussions 12.4.1 CAD system performance comparison with senior and trainee radiologist 12.4.2 Performance time of the developed CAD system 12.4.3 Application usability 12.4.3.1 System usability scale 12.4.3.2 Evaluation of GUI with SUS 12.4.4 Strengths, limitations, and future works 12.5 Conclusion Acknowledgments References 13. Optimization-based multilevel threshold image segmentation for identifying ischemic stroke lesion in brain MR images 13.1 Introduction 13.2 Methodology 13.2.1 MR image dataset 13.2.2 Preprocessing 13.2.3 Segmentation 13.2.3.1 Multilevel thresholding Between-class variance (Otsu's method) Kapur's method 13.2.3.2 Harmony search optimization algorithm 13.2.3.3 Electromagnetism–optimization algorithm 13.2.4 Performance evaluation of image segmentation 13.2.4.1 Probability rand index 13.2.4.2 Variation of information 13.2.4.3 Global consistency error 13.2.4.4 Structural similarity index 13.2.4.5 Feature similarity index metric 13.2.4.6 Matthews correlation coefficient 13.3 Results and discussion 13.4 Conclusion References 14. A study of machine learning algorithms used for detecting cognitive disorders associated with dyslexia 14.1 Introduction to neurological disorders 14.2 Classification of neurological disorders 14.2.1 Neurobiological disorder—an example: dyslexia 14.2.1.1 Causes 14.2.1.2 Symptoms 14.2.1.3 Detection methods 14.2.1.4 Challenges 14.2.1.5 Types of dyslexia 14.2.1.6 Intervention programs 14.2.2 Examples of neurocognitive disorder: dementia and Alzheimer's disease 14.2.2.1 Causes 14.2.2.2 Challenges 14.2.2.3 Dementia 14.2.2.4 Stages of dementia 14.2.2.5 Detection methods 14.2.3 Benchmarking digital neuroimaging resources 14.3 Machine learning algorithms 14.3.1 Supervised learning techniques 14.3.1.1 Association Rule Mining 14.3.1.2 Naive Bayes 14.3.1.3 Decision trees 14.3.1.4 Random forest 14.3.1.5 Neural networks and deep learning 14.3.1.6 Ensemble methods 14.3.1.7 Support vector machines 14.3.2 Unsupervised learning techniques 14.3.3 Statistical model 14.4 Conclusion References 15. A Critical Analysis and Review of Assistive Technology: Advancements, Laws, and Impact on Improving the Rehabilitation of D ... 15.1 Introduction 15.2 Dysarthria 15.3 Assistive technologies for dysarthria 15.3.1 Hard assistive technologies 15.3.2 Soft assistive technologies 15.4 Design considerations in the development of Assistive Technology 15.5 Assistive technology laws for improving the rehabilitation of dysarthric patients 15.5.1 United States 15.5.2 India 15.5.3 Europe 15.6 Impact of assistive technology on quality of life of dysarthric individuals 15.7 Inference 15.7.1 Advancements in assistive technology 15.7.2 Assistive technology laws 15.7.3 Impact of assistive technology 15.8 Summary of observations 15.9 Conclusion References 16. A comparative study on the application of machine learning algorithms for neurodegenerative disease prediction 16.1 Introduction 16.1.1 Alzheimer's statistics 16.2 Literature survey 16.3 Description of the dataset 16.4 Support vector machine 16.4.1 Algorithm 16.5 Decision tree 16.5.1 Algorithm 16.6 Random forest tree 16.6.1 Algorithm 16.7 Results and discussions 16.8 Conclusion and future enhancements References Index A B C D E F G H I K L M N O P Q R S T U V W