Explainable Machine Learning for Multimedia Based Healthcare Applications

✍ Scribed by M. Shamim Hossain (editor), Utku Kose (editor), Deepak Gupta (editor)

• Publisher: Springer
• Year: 2023
• Tongue: English
• Leaves: 240
• Category: Library

✦ Synopsis

This book covers the latest research studies regarding Explainable Machine Learning used in multimedia-based healthcare applications. In this context, the content includes not only introductions for applied research efforts but also theoretical touches and discussions targeting open problems as well as future insights. In detail, a comprehensive topic coverage is ensured by focusing on remarkable healthcare problems solved with Artificial Intelligence. Because today’s conditions in medical data processing are often associated with multimedia, the book considers research studies with especially multimedia data processing.

✦ Table of Contents

Foreword
Preface
Acknowledgement
Contents
Automatic Fetal Motion Detection from Trajectory of US Videos Based on YOLOv5 and LSTM
1 Introduction
2 Material and Method
2.1 Dataset
2.2 Structure of YOLO v5
2.3 LSTM (Long-Short Term Memory) Deep Neural Networks
3 Experimental Analysis
4 Conclusion
References
Explainable Machine Learning (XML) for Multimedia-Based Healthcare Systems: Opportunities, Challenges, Ethical and Future Pros...
1 Introduction
1.1 The Following Are the Significant Contributions of This Chapter
1.2 Chapter Organization
2 Multimedia Data in Healthcare Systems
2.1 Types of Multimedia Presentations in Healthcare Systems
2.1.1 Audio
2.1.2 Visual Data
2.1.3 Video
2.1.4 Text
3 Explainable Machine Learning for Multimedia Data in Healthcare Systems
3.1 A Classification of Techniques: Various Interpretability Scopes for Machine Learning
4 The Challenges of Explainable Machine Learning in Healthcare Systems
4.1 Lack of Standardized Requirements for XML
4.2 Unstandardized Representation Techniques
4.3 What Clinicians Expect: Explainability vs. Accuracy
4.4 What and How of the Results Explained
4.5 Security and Privacy Issues
4.6 Verification of Explanations
4.7 Ethical Restrictions
4.8 Lack of Theoretical Knowledge
4.9 Absence of Cause
5 An Effective Explainable Machine Learning Framework for Healthcare Systems
6 Research Prospects and Open Issues
7 Conclusion and Future Directions
References
Ensemble Deep Learning Architectures in Bone Cancer Detection Based on Medical Diagnosis in Explainable Artificial Intelligence
1 Introduction
2 Related Works
3 System Model
3.1 Optimized Kernel Fuzzy C Means Multilayer Deep Transfer Convolutional Learning (OpKFuzCMM-DTCL) Based Segmentation and Cla...
3.2 Performance Analysis
3.3 Dataset Description
3.4 Discussion
4 Conclusion
References
Digital Dermatitis Disease Classification Utilizing Visual Feature Extraction and Various Machine Learning Techniques by Expla...
1 Introduction
2 Materials and Methods
3 Results
4 Conclusion
References
Explainable Machine Learning in Healthcare
1 Introduction
2 Data Set Used
3 Various Machine Learning Algorithms
4 Linear Regression
5 SVM
6 Naive Bayes
7 Logistic Regression
8 K-Nearest Neighbors (kNN)
9 Decision Trees
10 RF Algorithm
11 Boosted Gradient Decision Trees (GBDT)
12 Clustering with K-Means
13 Analysis by Principal Components (PCA)
14 Case Study
15 Handling Missing Values
16 Result
17 Conclusion
18 Future Scope
References
Explainable Artificial Intelligence with Scaling Techniques to Classify Breast Cancer Images
1 Introduction
2 Related Work
3 Materials and Methods
3.1 Proposed Methodology
3.2 Dataset
3.3 Data Processing
3.3.1 Min-Max Scaling
3.3.2 Normalization
L1 Normalization
L2 Normalization
3.3.3 Z-score
3.4 Model
3.4.1 Logistic Regression
3.4.2 Support Vector Machine (SVM)
Support Vectors
Hyperplane
Margin
SVM Kernels
Linear Kernel
Polynomial Kernel
Radial Basis Function (RBF) Kernel
3.4.3 Decision Tree
Gini Index
Split Creation
3.4.4 Building a Tree
Terminal Node Creation
Recursive Splitting
3.4.5 Naïve Bayes
3.4.6 Random Forest
Working on Random Forest Algorithm
3.4.7 K-Nearest Neighbor (KNN)
3.4.8 Adaptive Boosting
3.4.9 Extreme Gradient Boosting
3.5 Performance Evaluation Metrics.
3.5.1 Confusion Matrix
3.5.2 Classification Report
Accuracy
Precision
Recall or Sensitivity
Specificity
F1 Score (F-measure)
Area Under ROC Curve (AUC)
3.5.3 Logarithmic Loss (LOGLOSS)
3.6 Metrics Use Case
3.7 Explainable Artificial Intelligence (XAI)
4 Result and Discussion
4.1 Experimental Setup
4.2 Explainable Result
4.3 Experimental Results
4.3.1 SVM
Random Forest (RF)
4.3.2 Naive Bayes (NB)
4.3.3 Logistic Regression (LR)
4.3.4 KNN
4.3.5 Decision Tree (DT)
4.3.6 Adaboost
4.3.7 XGBoost
4.3.8 Comparative Analysis
5 Conclusion and Future Work
References
A Novel Approach of COVID-19 Estimation Using GIS and Kmeans Clustering: A Case of GEOAI
1 Introduction
2 Discussion and Results
2.1 Temporal Data Distribution
2.2 Equation Based on the Values of the Confirmed Cases Obtained
2.3 Artificial Intelligence in Covid-19 Drones for Survey During Lockdown
2.4 Robots During Covid-19
3 Conclusions
References
A Brief Review of Explainable Artificial Intelligence Reviews and Methods
1 Introduction
2 Fundamental Definitions
3 Recent XAI Methods and Reviews
3.1 Review Studies
3.2 XAI Methods
4 XAI in Medicine
4.1 SHAP
4.2 GRADCAM
4.3 LRP
4.4 Lime
5 Discussion and Future Directions
6 Conclusion
References
Systematic Literature Review in Using Big Data Analytics and XAI Applications in Medical
1 Introduction
1.1 The Concept of Big Data
1.2 Explainable Artificial Intelligence: XAI
1.3 Big Data Analytics and XAI Applications in Medical
2 Methodology
2.1 Research Questions
2.2 Research Process
2.3 Data Collection
2.4 Data Analysis
3 Discussion and Results
3.1 Context Results
3.2 Evaluation of the Studies
4 Conclusions
References
Using Explainable Artificial Intelligence in Drug Discovery: A Theoretical Research
1 Introduction
1.1 A General Introduction to the Drug Discovery Sector
2 Stages of Drug Discovery
2.1 Using Artificial Intelligence in Drug Discovery Phases
3 What Is Explainable Artificial Intelligence?
3.1 The Importance of Explainable Artificial Intelligence in Drug Discovery
4 Academic Studies in Drug Discovery
5 Conclusions
References
Application of Interpretable Artificial Intelligence Enabled Cognitive Internet of Things for COVID-19 Pandemics
1 Introduction
2 Applications of Explainable Artificial Intelligence in COVID-19 Pandemics
3 Cognitive Internet of Things for COVID-19 Pandemics
3.1 Rapid Diagnosis
3.2 Contact Tracing and Clustering
3.3 Prevention and Control
3.4 Screening and Surveillance
3.5 Remote Monitoring of the Patient
3.6 Real-Time Tracking
3.7 Development of Drugs and Vaccines
4 The Challenges of Interpretable Artificial Intelligence Enabled Cognitive Internet of Things for COVID-19 Pandemic
5 The Framework of an XAI Enabled CIoT for Fighting COVID-19 Pandemic
6 Conclusion and Future Directions
References
Remote Photoplethysmography: Digital Disruption in Health Vital Acquisition
1 Introduction
2 Basic Principle
2.1 Photoplethysmography (PPG)
2.2 Remote Photoplethysmography (rPPG)
2.2.1 Principle of rPPG
2.2.2 Skin Reflection Model
2.2.3 Use of AI in rPPG
3 Algorithmic Methods
3.1 Blind Source Separation (BSS) Method (PCA/ICA)
3.2 Model-Based Method (CHROM/BVP)
3.2.1 Chrominance-Based Method (CHROM)
3.2.2 BVP Signature-Based Method
3.3 Design-Based Method
4 Issues and Literature Review
4.1 PPG vs. rPPG
4.2 Factors Affecting rPPG Video Capturing
4.2.1 Effect of Light Source
4.2.2 Effect of Body Motion
4.2.3 Effect of Camera´s Frame Rate
4.3 Effect of Video Compression
4.4 ROI Detection and Selection Problem
4.5 Signal Processing Techniques Limitations
4.6 Extracted Signal Noise Problem
5 Trends and Tools
6 Study and Results
7 Conclusion
Annexure-I: Informed Consent Form
References