Advanced Machine Learning Approaches in Cancer Prognosis: Challenges and Applications (Intelligent Systems Reference Library, 204)

✍ Scribed by Janmenjoy Nayak (editor), Margarita N. Favorskaya (editor), Seema Jain (editor), Bighnaraj Naik (editor), Manohar Mishra (editor)

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 461
• Edition: 1st ed. 2021
• Category: Library

✦ Synopsis

This book introduces a variety of advanced machine learning approaches covering the areas of neural networks, fuzzy logic, and hybrid intelligent systems for the determination and diagnosis of cancer. Moreover, the tactical solutions of machine learning have proved its vast range of significance and, provided novel solutions in the medical field for the diagnosis of disease. This book also explores the distinct deep learning approaches that are capable of yielding more accurate outcomes for the diagnosis of cancer. In addition to providing an overview of the emerging machine and deep learning approaches, it also enlightens an insight on how to evaluate the efficiency and appropriateness of such techniques and analysis of cancer data used in the cancer diagnosis. Therefore, this book focuses on the recent advancements in the machine learning and deep learning approaches used in the diagnosis of different types of cancer along with their research challenges and future directions for the targeted audience including scientists, experts, Ph.D. students, postdocs, and anyone interested in the subjects discussed.

✦ Table of Contents

Foreword
Preface
Contents
About the Editors
Part I Cancer Data Analysis with Machine Learning Approaches
1 Advances in Machine Learning Approaches in Cancer Prognosis
1.1 Introduction
1.2 Key Predictive Challenges
1.2.1 Machine Learning Methods for Cancer Susceptibility
1.2.2 Machine Learning Methods for Cancer Recurrence
1.2.3 Machine Learning Methods for Cancer Outcomes
1.3 Conclusions
References
2 Data Analysis on Cancer Disease Using Machine Learning Techniques
2.1 Introduction
2.1.1 Overview of the Cancer Disease
2.1.2 Overview of the Cancer Data
2.1.3 Objective and Proposed Outcomes
2.1.4 Organization of the Chapter
2.2 Review Report
2.2.1 Supervised Learning
2.2.2 Semi-supervised Learning
2.2.3 Unsupervised Learning
2.3 Challenges and Issues in Data Analysis of Cancer Disease
2.4 Comparative Study Through Validation Parameters
2.5 Comparative Study of Data Analysis in Cancer Disease
2.6 Future Scope
2.7 Discussion
2.8 Conclusion
References
3 Learning from Multiple Modalities of Imaging Data for Cancer Detection/Diagnosis
3.1 Introduction
3.2 Development of a Method to Visualization of Multispectral Endoscopic Images
3.2.1 The Main Idea
3.2.2 Methods and Algorithms for Endoscopic Images Enhancement
3.2.3 Objective Metrics for Assessing Image Quality Enhancement
3.2.4 Experimental Results
3.2.5 Superimposing Color and NIR Channels in Laparoscopic Imaging
3.3 Anatomical Landmarks Detection for Laparoscopic Surgery Based on Deep Learning
3.4 Discussion
3.5 Conclusion
References
4 Neural Network for Lung Cancer Diagnosis
4.1 The ‘AI’ Motto
4.2 Limitations of AI/ML
4.3 Deep Learning in Medical Imaging
4.4 Neural Networks for Lung Cancer Diagnosis
4.4.1 Convolution Neural Networks
4.5 Proposing a 3D Neural Network
4.6 Applying the Concept of Short-Breaths and Thermal Imaging
4.6.1 Components and Arrangements
4.7 Contemplation
References
5 Improved Thyroid Disease Prediction Model Using Data Mining Techniques with Outlier Detection
5.1 Introduction
5.2 Data Mining and Medical Data Mining
5.3 Related Work
5.4 Methodology
5.4.1 Selecting Database
5.4.2 Tools Used
5.4.3 Major Classification Algorithms
5.5 Experimental Setup
5.5.1 Tools Used
5.5.2 Performance Evaluation Strategies
5.5.3 10 Fold Cross Validation
5.5.4 Experimental Results
5.6 Conclusion
5.7 Future Work
References
6 Automated Breast Cancer Diagnosis Based on Neural Network Algorithms
6.1 Introduction
6.1.1 Benign (Non-cancerous)
6.1.2 Malignant
6.2 Related Work
6.2.1 CAD (Computer Aided Diagnosis)
6.2.2 Feature Extraction
6.2.3 Gray-Map
6.2.4 SOBEL
6.2.5 SGLDM
6.2.6 AFUM
6.2.7 SFUM
6.2.8 Deep Learning Techniques
6.3 Methodology
6.3.1 Histogram Equalization Used in Proposed System
6.3.2 Adaptive Median Filter
6.3.3 Hardware Implementation of Filter
6.3.4 Gaussian Mixture Models
6.3.5 Gray Level Co-occurrence Matrix
6.3.6 Probabilistic Neural Network
6.3.7 Working of PNN
6.3.8 Pseudo Code for Probabilistic Neural Network
6.3.9 Experimental Setup
6.3.10 Dataset
6.4 Results
6.5 Conclusion
References
Part II Prediction of Cancer Susceptibility
7 Feature Extraction and Classification of Colon Cancer Using a Hybrid Approach of Supervised and Unsupervised Learning
7.1 Introduction
7.2 Realted Work
7.3 Contribution
7.4 Summary
7.5 Material and Proposed Methodology
7.5.1 Dataset
7.5.2 Pre-processing : Clustering
7.5.3 Clustering Validation
7.5.4 Oversampling: Adaptive Synthetic Sampling
7.5.5 Feature Selection
7.5.6 Classification
7.6 Experimental Results
7.6.1 Statistical Analysis
7.6.2 Performance Comparison Using Receiver Operating Characteristic Curve
7.7 Conclusion
References
8 Automatic Detection of Tumor Cell in Brain MRI Using Rough-Fuzzy Feature Selection with Support Vector Machine and Morphological Operation
8.1 Introduction
8.2 Literature Study
8.3 Proposed Method
8.3.1 Data Set Creation
8.3.2 Feature Set Extraction
8.3.3 Feature Selection Using Rough Kernelized Fuzzy C-Means Clustering
8.3.4 Tumor Classification Using SVM
8.3.5 Tumor Detection and Extraction
8.4 Experimental Results
8.5 Conclusion
References
9 Overlapping Oral Epithelial Cells Segmentation: Voronoi-Based Hybrid Active Contour Model
9.1 Introduction
9.2 Related Works
9.2.1 Cytological Analysis on Oral Cancer
9.2.2 Related Works on This Present Work
9.3 Materials
9.3.1 Sample Collection and Preparation
9.3.2 Image Acquisition from Microscopic Slides and Generation of Database
9.4 Proposed Method
9.4.1 Pre-processing
9.4.2 Initial Segmentation
9.4.3 Debris Removal
9.4.4 Nucleus Segmentation
9.4.5 Overlapping Cell Segmentation
9.5 Experimental Results
9.5.1 Performance Evaluation
9.5.2 Experimental Results of Overall Cell Segmentation
9.5.3 Segmentation Performance Assessment
9.6 Summary
References
10 In Silico Modeling of Anticancer Drugs: Recent Advances
10.1 Introduction
10.2 Drug Design for Anti-cancer Drugs
10.2.1 Drug Discovery Cycle
10.2.2 In Silico Approaches for Drug Design
10.2.3 Structure-Based Drug Design
10.2.4 Virtual Screening
10.3 Main Approaches to Design Anti-cancer Drugs
10.4 Some Important Examples of In-Silico Design in Anti-cancer Drugs
10.5 Conclusion
References
11 Two Dimensional and Gesture Based Medical Visualization Interface and Image Processing Methodologies to Aid and Diagnose of Lung Cancer
11.1 Introduction
11.2 Image Processing Methodologies for Lung Cancer
11.3 Proposed Methodology
11.3.1 Overview
11.3.2 Image Enhancement
11.4 Results and Discussion
11.4.1 Selection Gesture
11.4.2 Swipe Gesture
11.4.3 Drag Gesture
11.5 Conclusion
References
Part III Advanced Machine Learning Paradigms for Cancer Diagnosis
12 Deep MammoNet: Early Diagnosis of Breast Cancer Using Multi-layer Hierarchical Features of Deep Transfer Learned Convolutional Neural Network
12.1 Introductions
12.2 Literature Study
12.2.1 In Reviews & Surveys
12.2.2 In Machine Learning
12.2.3 In Deep Learning
12.2.4 Problem Statement
12.2.5 Contribution
12.3 Proposed Method
12.3.1 Hierarchical Features
12.3.2 Mammogram Augmentation
12.4 Experimental Analysis and Discussion
12.4.1 Corpus
12.4.2 Preprocessing
12.4.3 Analysis and Discussion
12.5 Conclusion and Future Scope
References
13 Study on Gene Alterations in Cervical Cancer Using Computational Genomics Tools
13.1 Introduction
13.1.1 Gene Expression Omnibus (GEO)
13.1.2 NetworkAnalyst
13.1.3 GeneMANIA
13.1.4 GeneCards
13.1.5 cBioPortal
13.2 Methodology
13.2.1 NCBI Geo
13.2.2 Data Analysis in NetworkAnalyst
13.2.3 Identifying Co-expressed Genes Using GeneMANIA
13.2.4 Summary of Genes Using GeneCards
13.2.5 Gene Alterations Using cBioPortal
13.3 Results
13.4 Conclusion
References
14 Prostate Cancer: Cancer Detection and Classification Using Deep Learning
14.1 Introduction
14.2 Dataset
14.3 Methodology
14.3.1 Data Pre-processing
14.3.2 Deep Learning (DL) Models
14.3.3 Evaluation Metrics
14.4 Results and Discussions
14.5 Conclusion
References
15 A Deep Learning Prediction Model for Detection of Cancerous Lesions from Dermatoscopic Images
15.1 Introduction
15.2 Literature Study
15.3 Preliminaries, Tools and Techniques
15.3.1 Convolutional Neural Network (CNN)
15.3.2 Adams Optimizer
15.3.3 Categorical Cross Entropy
15.3.4 Softmax
15.4 Dataset
15.4.1 ISBI Challenge Dataset
15.4.2 PH2 Dataset
15.4.3 ISIC Challenge
15.4.4 Kaggle
15.5 Proposed Methodology
15.5.1 Training Phase
15.5.2 Testing Phase
15.5.3 Proposed Architecture
15.5.4 Training and Testing Procedure
15.6 Results
15.6.1 Accuracy
15.6.2 Comparison with Existing Works
15.7 Conclusions
References
16 Deep Learning Based Classification of Brain Tumor Types from MRI Scans
16.1 Introduction
16.2 Literature Study
16.3 Methodology Used
16.3.1 Convolutional Neural Network (CNN)
16.3.2 Evaluating the Performance of a Classification Model
16.4 Proposed Method
16.4.1 Proposed Model Architecture
16.4.2 Modelling the Classification Task
16.5 Experimental Setup
16.5.1 Dataset Preparation and Pre-processing
16.5.2 Data Partitioning
16.5.3 Training and Testing
16.6 Results, Analysis and Discussions
16.6.1 Comparison with Existing Works
16.7 Conclusion
References