Cognitive and Soft Computing Techniques for the Analysis of Healthcare Data

✍ Scribed by Akash Kumar Bhoi, Victor Hugo Costa de Albuquerque, Parvathaneni Naga Srinivasu, Goncalo Marques

• Publisher: Academic Press
• Year: 2022
• Tongue: English
• Leaves: 278
• Series: Intelligent Data-Centric Systems
• Category: Library

✦ Synopsis

Cognitive and Soft Computing Techniques for the Analysis of Healthcare Data discusses the insight of data processing applications in various domains through soft computing techniques and enormous advancements in the field.

The book focuses on the cross-disciplinary mechanisms and ground-breaking research ideas on novel techniques and data processing approaches in handling structured and unstructured healthcare data. It also gives insight into various information-processing models and many memories associated with it while processing the information for forecasting future trends and decision making.

This book is an excellent resource for researchers and professionals who work in the Healthcare Industry, Data Science, and Machine learning.

✦ Table of Contents

Cognitive and Soft Computing Techniques for the Analysis of Healthcare Data
Copyright
Contributors
Preface
1 . Artificial intelligence and machine learning for the healthcare sector: performing predictions and metrics evaluation of ML ...
1. Introduction
1.1 Artificial intelligence and health care
1.2 Opportunities and challenges of AI and ML in the healthcare sector
2. Smart healthcare system
3. Machine learning example of data analytics in health care
3.1 Data classification
3.1.1 Data collection
3.1.2 Data preprocessing
3.1.3 ML algorithms applied
3.1.3.1 Logistic regression
3.1.3.2 Support vector machines
3.1.3.3 Boosted and bagged trees
3.1.3.4 Naïve Bayes
3.1.3.5 Decision tree
3.1.4 Training and testing
3.2 Clustering techniques
4. Experimental results
5. Conclusion
Abbreviations
References
2 . Cognitive technology for a personalized seizure predictive and healthcare analytic device
1. Introduction
2. Epilepsy and seizures
2.1 Quick statistics
2.2 Types of seizures
2.2.1 Generalized-onset seizures
2.2.2 Focal-onset seizures
3. Cognitive technology
4. Internet of Things
4.1 Why do we need the IoT?
4.2 Working principle
5. Cognitive IoT and neural networks
5.1 Deep neural networks
5.2 Steps to construct an artificial neural network or deep neural network
6. Natural language processing
6.1 Classical models
6.2 Deep learning models
6.3 Applications of natural language processing
7. Problem statement
8. Methodology
9. Proposed approach
9.1 Working of the system
9.2 Device for different types of seizures
9.3 Device for health monitoring
10. Simulations and discussions
11. Conclusions
References
3 . Cognitive Internet of Things (IoT) and computational intelligence for mental well-being
1. Introduction
2. Cognitive IoT and computational intelligence in health care
3. Computer vision for early diagnosis of mental disorders using MRI
4. Feature selection techniques and optimization techniques used
5. Natural language processing-based diagnostic system
6. Harnessing the power of NLP for the analysis of social media content for depression detection
7. Computational intelligence and cognitive IoT in suicide prevention
8. Wearables and IoT devices for mental well-being
9. Future scope of computational intelligence in mental well-being
10. Conclusion
References
4 . Artificial neural network-based approaches for computer-aided disease diagnosis and treatment
1. Introduction
1.1 Structure of this chapter
2. Artificial neural networks applied to computer-aided diagnosis and treatment
2.1 Computer-aided diagnosis and treatment systems (CADTS)
2.1.1 Challenges in CADTS
2.2 Artificial neural networks in medical diagnosis
2.3 ANN: types and applications
2.3.1 Feed-forward neural networks
2.3.2 Recurrent neural networks
2.3.3 Deep learning convolutional neural networks
2.4 Future trends in ANN for disease diagnosis and treatment
3. Application of ANN in the diagnosis and treatment of cardiovascular diseases
3.1 Applications in cardiology
3.1.1 Echocardiography
3.1.2 Electrocardiography
3.1.3 Interventional cardiology: angiograms
4. Case study: ANN and medical imaging—brain tumor detection
4.1 Proposed methodology
4.1.1 Convolutional neural networks
4.1.2 Transfer learning
4.1.3 VGG16
4.1.4 Proposed deep learning architecture
4.1.5 Data set
4.2 Model analysis
4.2.1 Confusion matrix
4.2.2 Precision
4.2.3 Recall
4.2.4 Accuracy
4.3 Experiments and results
5. Final considerations
References
5 . AI and deep learning for processing the huge amount of patient-centric data that assist in clinical decisions
1. Introduction
1.1 Structure of this chapter
2. Challenges and trends
2.1 Clinical decision support systems (CDSSs)
2.2 Artificial intelligence × knowledge base systems
2.3 The adoption of CDSSs
2.4 The challenge of the increasing amounts of data available for clinical decision-making
2.4.1 Big data requirements and developing countries
2.5 Artificial intelligence and deep learning for CDSS
2.6 Engineers and medical researchers
2.7 Developments and trends in big data and AI for CDSS
3. Case study 1: multiple Internet of Things (IoT) monitoring systems and deep learning classification systems to support ambu ...
3.1 Preliminary CDSS
3.2 Deep learning classifier
3.3 Conclusion
4. Case study 2: artificial intelligence epidemiology prediction system during the COVID-19 pandemic to assist in clinical dec ...
4.1 Long-term short-term memory networks (LSTM)
4.2 Auto machine learning approach (AutoML)
4.3 Results and discussion
4.3.1 LSTM predictions for Brazil
4.3.2 H2O AutoML results for Brazil
4.4 Conclusion
5. Final considerations
References
6 . Universal intraensemble method using nonlinear AI techniques for regression modeling of small medical data sets
1. Introduction and problem statement
2. Related concepts
3. Universal intraensemble method for handling small medical data
3.1 Design of the universal intraensemble method
3.2 Algorithm 1: support vector regression using nonlinear kernels
3.3 Algorithm 2: general regression neural network
3.4 Algorithm 3: RBF neural network
4. Practical implementation
4.1 Data sets for experimental modeling
4.2 Performance indicators
4.3 Optimal parameters selection
4.4 Algorithm 1: support vector regression using nonlinear kernels
4.5 Algorithm 2: general regression neural network
4.6 Algorithm 3: RBF neural network
4.7 Results
5. Comparison and discussion
5.1 A comparative study of three different algorithms
5.2 Comparison with parental regressors
6. Conclusion and future work
Appendix A
References
7 . Comparisons among different stochastic selections of activation layers for convolutional neural networks for health care
1. Introduction
2. Literature review
3. Activation functions
4. Materials and methods
5. Results
6. Conclusions
Acknowledgments
References
8 . Natural computing and unsupervised learning methods in smart healthcare data-centric operations
1. Introduction
2. Natural computing in the healthcare industry
2.1 Computing inspired by nature (CIN)
2.2 Recent works using NC algorithms in solving various problems in healthcare systems
3. Unsupervised learning techniques in healthcare systems
3.1 Clustering
3.2 Association
4. The data-centric operations in healthcare systems
4.1 Applications of data-centric intelligence to healthcare systems
5. Case study for application of the particle swarm optimization model for the diagnosis of heart disease
5.1 The heart disease data set characteristics
5.2 Performance evaluation metrics
6. Results and discussion
7. Conclusion
References
9 . Optimized adaptive tree seed Kalman filter for a diabetes recommendation system—bilevel performance improvement strategy fo ...
1. Introduction
2. Literature review
3. The proposed AKF-TSA-based insulin recommendation system
3.1 Kalman filtering technique
3.2 The adaptive Kalman filtering (AKF) technique
3.3 Tree seeding optimization algorithm
3.4 Recommendation process
4. Results and discussion
4.1 Data set used
4.2 Performance validation
5. Conclusion
References
10 . Unsupervised deep learning-based disease diagnosis using medical images
1. Introduction
2. Related works
3. Methodology
3.1 Feature extraction using PCA-Net
3.1.1 First convolution stage
3.1.2 Second convolution stage
3.1.3 The output stage
3.2 Working of principal component analysis (PCA) operation
3.3 K-means classifier
4. Experiments
4.1 Data set
4.2 Hyperparameters
4.3 Incremental training of K-means
5. Evaluation metrics
5.1 Accuracy
5.2 Precision
5.3 Recall/sensitivity
5.4 F1-score
5.5 Specificity
5.6 Matthews correlation coefficient
5.7 Receiver operating characteristic (ROC) curve
5.8 Area under the ROC curve (AUC)
6. Experimental results and discussions
7. Conclusion
8. Future work
References
11 . Probabilistic approaches for minimizing the healthcare diagnosis cost through data-centric operations
1. Introduction
2. Bayesian neural networks
3. Markov chain Monte Carlo (MCMC)
3.1 Variational inference (VI)
3.2 Applications
4. Breast cancer prediction using a Bayesian neural network
5. Conclusion
References
12 . Effects of EEG-sleep irregularities and its behavioral aspects: review and analysis
1. Introduction
2. Medical background
3. Visual scoring procedure
4. AI and sleep staging
4.1 Analysis of human sleep behavior using sleep variables
4.2 Characteristics of biosignals
4.2.1 Technical characteristics
4.2.2 Clinical characteristics
4.3 Non-REM sleep
4.4 Sleep diaries and questions
5. Sleep patterns and clinical age
5.1 Newborns
5.2 Young children
5.3 Adolescents
5.4 Adults
5.5 Gender differences
5.6 Elderly people
6. Case study of an automated sleep staging system
6.1 Experimental data
6.1.1 Sleep-EDF (S-EDF) database
6.2 Proposed ensemble learning stacking model
6.3 Sleep staging results in the S-EDF database
7. Chapter outcome and conclusion
References
Index
A
B
C
D
E
F
G
H
I
K
L
M
N
O
P
R
S
T
U
V
W