The book discusses Explainable (XAI) and Responsive Artificial Intelligence (RAI) for biomedical and healthcare applications. It will discuss the advantages in dealing with big and complex data by using explainable AI concepts in the field of biomedical sciences. The book explains both positive as well as negative findings obtained by explainable AI techniques. It features real time experiences by physicians and medical staff for applied deep learning based solutions. The book will be extremely useful for researchers and practitioners in advancing their studies.
Author(s): Aditya Khamparia, Deepak Gupta, Ashish Khanna, Valentina E. Balas
Series: Intelligent Systems Reference Library, 222
Publisher: Springer
Year: 2022
Language: English
Pages: 147
City: Singapore
Preface
Contents
Editors and Contributors
1 Optimal Boosting Label Weighting Extreme Learning Machine for Mental Disorder Prediction and Classification
1.1 Introduction
1.2 The Proposed Model
1.2.1 Data Pre-processing
1.2.2 Process Involved in BWELM Model
1.2.3 Parameter Tuning Using CSPSO Algorithm
1.3 Experimental Validation
1.4 Conclusion
References
2 Modeling of Explainable Artificial Intelligence with Correlation-Based Feature Selection Approach for Biomedical Data Analysis
2.1 Introduction
2.2 The Proposed Model
2.2.1 Stage 1: Pre-processing
2.2.2 Stage 2: Correlation-Based Feature Selection
2.2.3 Stage 3: FKNN-Based Classification
2.2.4 Stage 4: BWO-Based Classification
2.3 Experimental Validation
2.4 Conclusion
References
3 Explainable Machine Learning Model for Diagnosis of Parkinson Disorder
3.1 Introduction
3.2 Literature Survey
3.3 Experimental Work
3.4 Discussion
3.5 Conclusion
References
4 Explainable Artificial Intelligence with Metaheuristic Feature Selection Technique for Biomedical Data Classification
4.1 Introduction
4.2 Literature Review
4.3 The Proposed Model
4.3.1 Data Preprocessing
4.3.2 Algorithmic Design of CSMO Based Feature Selection
4.3.3 Process Involved in Optimal DNN-Based Classification
4.4 Experimental Validation
4.5 Conclusion
References
5 Explainable AI in Neural Networks Using Shapley Values
5.1 Introduction
5.2 Literature Review
5.2.1 Explanation by Simplification
5.2.2 Explanation by Feature Attribution
5.3 Shapley Values and Game Theory
5.3.1 Using Shapley Values to Attributing Relevance
5.3.2 Shapley Value to SHAP
5.4 Explainer Architecture
5.4.1 Model Explainer
5.4.2 Visualization
5.4.3 Model Refinement
5.4.4 Reporting and Presentation
5.5 Discussion
5.5.1 Comparison with Other Explainable Methods
5.5.2 Axiomatic Comparison
5.6 Conclusion and Future Work
References
6 Design of Multimodal Fusion-Based Deep Learning Approach for COVID-19 Diagnosis Using Chest X-Ray Images
6.1 Introduction
6.2 Literature Survey
6.3 The Proposed MMFBDL Model
6.3.1 Feature Extraction Process
6.3.2 Image Classification Using MLP
6.4 Experimental Validation
6.5 Conclusion
References
7 ECG Classification and Analysis for Heart Disease Prediction Using XAI-Driven Machine Learning Algorithms
7.1 Introduction
7.2 Literature Review
7.3 Dataset Methods and Classification
7.3.1 Tools and Techniques
7.3.2 ECG Results Implementation for Normal and Abnormal
7.3.3 Results for Individual Disease by Cross-Validation Score
7.3.4 Cross-Validation Score for ANN
7.4 Description of the ML Models
7.4.1 Logistic Regression
7.4.2 Naive Bayes
7.4.3 Decision Trees
7.4.4 Support Vector Machine (SVM)
7.4.5 Lime
7.4.6 DeepLIFT
7.4.7 Skater
7.4.8 Shapley
7.5 Results Analysis
7.6 Conclusion and Future Scope
References
8 Rethinking the Transfer Learning Architecture for Respiratory Diseases and COVID-19 Diagnosis
8.1 Introduction
8.2 Literature Reviews
8.3 Description of Dataset
8.4 Methodology
8.4.1 VGG-16
8.4.2 XceptionNet Model
8.5 Result Analysis
8.6 Conclusion
References
9 Arithmetic Optimization Algorithm with Explainable Artificial Intelligence Technique for Biomedical Signal Analysis
9.1 Introduction
9.2 Related Works
9.3 The Proposed Model
9.3.1 Variation Mode Decomposition (VMD) Approach
9.3.2 Feature Extraction Using Bi-LSTM Model
9.3.3 ECG Recognition Using Optimal SVM Model
9.4 Experimental Validation
9.5 Conclusion
References
