Deep Learning for Biomedical Applications

Cover
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
Foreword
Preface
Editors
Contributors
Chapter 1 Precision Medicine and Omics in the Context of Deep Learning
1.1 Introduction
1.2 Deep Learning Techniques and Methods
1.2.1 Deep Neural Networks
1.2.2 Convolutional Neural Network
1.2.3 Recurrent Neural Network
1.2.4 Long-/Short-Term Memory
1.2.5 Deep Belief Network
1.2.6 Autoencoder
1.3 Using Deep Learning to Analyze Omics Data
1.3.1 DL in Genomics
1.3.2 DL in Genome-Wide Association Studies
1.3.3 DL in Transcriptomics
1.3.4 DL in Epigenomics
1.3.5 DL in Proteomics and Metabolomics
1.4 Deep Learning Applications in Precision Medicine
1.4.1 Finding Biomarker and Classifying patients
1.4.2 Medical Imaging
1.4.3 Diagnosing Cancer
1.5 Conclusions
References
Chapter 2 Embryo Grade Prediction for In-Vitro Fertilization
2.1 Introduction
2.2 IVF Procedure
2.3 Embryo Assessment
2.4 CNN and Its Application to Embryo Grading
2.5 Joint Optimization of Image Processing Filter Selection and CNN Weights
2.5.1 Image Processing Filters
2.5.1.1 Median Filter
2.5.1.2 Gaussian Filter
2.5.1.3 Sobel Filter
2.5.1.4 Laplacian Filter
2.5.1.5 Sharpening Filter
2.5.1.6 Blur Filter
2.5.2 Proposed Algorithm
2.6 Experiment and Results
2.6.1 Data Set
2.6.2 Experimental Settings
2.6.3 Results and Discussion
2.7 Conclusion
Note
References
Chapter 3 Biometric Gait Features Analysis Using Deep Learning Approaches
3.1 Introduction: Background
3.2 Literature Review
3.3 Methodology Proposed and Adopted
3.3.1 Acquiring Information for Depth Video
3.3.2 Calculation and Extraction of Regions of Interest
3.3.3 Extracting Features from Acquired Depth Images from Kinect V2.0 Sensors
3.4 Results Analysis and Discussion
3.5 Conclusion and Future Work
3.6 Effects of This Research in Biomedical Domain
Acknowledgments
References
Chapter 4 Segmentation of Magnetic Resonance Brain Images Using 3D Convolution Neural Network
4.1 Introduction
4.2 Materials and Methods
4.2.1 Pre-processing
4.2.2 3D CNN Architecture
4.3 Results and Discussion
4.4 Inferences and Future Works
4.5 Conclusion
Acknowledgments
References
Chapter 5 Performance Analysis of Deep Learning Models for Biomedical Image Segmentation
5.1 Introduction
5.2 Methodology
5.3 Data Set Description
5.3.1 BraTs 2017 Brain Tumor
5.3.2 Preprocessed BraTs 2015 Brain Tumor
5.3.3 ISIC 2018 Skin Tumor
5.4 Architecture
5.4.1 Segmentor Adversarial Network
5.4.2 SegNet Architecture
5.4.3 U-Net Architecture
5.5 Evaluation Parameters
5.6 Results and Discussion
5.6.1 Comparison with Existing Approaches
5.7 Discussion
5.8 Conclusion
References
Chapter 6 Deep Learning for Ophthalmological Images
6.1 Introduction
6.2 Retinal Imaging Techniques
6.2.1 Fundus Fluorescein Angiography
6.2.2 Optical Coherence Tomography
6.2.3 Comparison of Fundus and OCT
6.3 Deep Learning
6.3.1 Convolutional Neural Network and Transfer Learning Architectures
6.3.2 Transfer Learning
6.4 Literature Review
6.5 Conclusion
References
Chapter 7 Deep Learning vs. Super Pixel Classification for Breast Masses Segmentation
7.1 Introduction
7.2 Background
7.2.1 Conventional Segmentation Methods
7.2.2 Model-Based Segmentation Methods
7.2.3 Deep Learning Techniques
7.3 Related Work
7.4 Methods
7.4.1 Super-Pixel-Based Segmentation
7.4.2 Deep Learning Approach
7.4.3 Experimental Setting
7.5 Results
7.5.1 Super-Pixels-Based Segmentation
7.5.2 Deep Learning Models
7.6 General Synthesis
7.7 Conclusions and Future Work
Acknowledgments
References
Chapter 8 Deep Learning for Disease Prediction in Public Health
8.1 Introduction
8.2 Big Public Health Data
8.3 Artificial Intelligence in Public Health
8.4 Deep Learning Approaches in Non-Communicable Diseases Prediction
8.4.1 Diabetes
8.4.2 Hypertension
8.4.3 Obesity
8.4.4 Predicting Hypertension, Hyperglycemia and Dyslipidemia
8.4.5 Cardiovascular Disease
8.5 Deep Learning Approaches in Communicable Diseases Prediction
8.5.1 COVID-19 Pandemic
8.5.2 Dengue Outbreak
8.6 Deep Learning Approaches in Socio-Behavioral Medicine
8.6.1 Predicting Mental Health States through Mobile Phones Metadata
8.6.2 Predicting Mental Health States through Social Media Data
8.6.3 Public Reactions and Responses During Outbreaks
8.7 Deep Learning Approaches in Maternal and Child Health Disease Prediction
8.7.1 Maternal Health
8.7.2 Child Health
8.8 Deep Learning Approaches in Environmental Health Disease Prediction
8.9 Strengths, Limitations and Future Direction
8.10 Conclusion
Acknowledgment
References
Chapter 9 Genomics with Deep Learning
9.1 Introduction: Background and Driving Forces
9.1.1 Genomics Primer
9.1.1.1 High-Throughput Sequencing
9.1.1.2 Sequencing Data: DNA, RNA and Protein
9.1.1.3 DNA Regulatory
9.1.1.4 Genetic Variation
9.1.2 Deep Learning Primer
9.1.3 Classification Applications: Deciphering Syntax of a Regulatory DNA Sequence
9.1.4 Regression Application: Gene Expression Level Prediction
9.2 Genomic Deep Learning Setup
9.2.1 Data Collection
9.2.2 Data Quality Control
9.2.3 Data Cleaning
9.2.4 Data Processing
9.2.5 Deep Learning Resources (Platform), Infrastructure, and Libraries
9.2.6 Deep Modelling from a Genomic Perspective
9.2.6.1 Convolutional Neural Network
9.2.6.2 Recurrent Neural Networks
9.2.6.3 Autoencoders
9.2.6.4 Emergent Architectures
9.2.7 Transfer Learning
9.2.8 Multi-Task and Multi-View Learning
9.3 Deep Learning Effectively on Genomic Data
9.3.1 Strict Data Set Hygiene
9.3.2 Always Be Close to Your Data!
9.3.3 Define Metrics and Establish a Baseline Model
9.3.4 Start Small, Iteratively Add Complexity
9.4 A Collection State-of-the-Art Models and Their Genomic Applications
9.4.1 SPROUT: CRISPR Outcome Prediction
9.4.2 Deep Learning for Rare Disease Diagnostics
9.4.3 Generative Models
9.4.3.1 Deep RNN Can be Used for Molecule Design
9.4.3.2 Deep Learning for Generate Anti-Microbial Peptide
9.4.3.3 GAN for DNA Optimizes Protein Function
9.5 Outro
9.5.1 Challenges of Deep Learning
9.5.2 Deep Learning Resources
9.5.3 Outlook
References
Chapter 10 A Review of Deep Learning-Based Methods for Cancer Detection and Classification
10.1 Introduction
10.2 Types of Cancer
10.2.1 Brain Cancer
10.2.2 Breast Cancer
10.2.3 Colon Cancer
10.2.4 Lung Cancer
10.2.5 Prostate Cancer
10.2.6 Skin Cancer
10.3 Deep Learning Architectures and Models for Cancer Detection and Classification
10.4 Data Sets for Cancer Detection and Classification Using Deep Learning Methods
10.4.1 MITOS-ATYPIA Data Set
10.4.2 BreakHis Data Set
10.4.3 INbreast Database
10.4.4 CBIS-DDSM Data Set
10.4.5 LIDC-IDRI Data Set
10.4.6 LUNA Data Set
10.4.7 DLCST Data Set
10.4.8 JSTR Data Set
10.4.9 HAM10000 Data Set
10.4.10 MED-NODE Data Set
10.4.11 PROMISE 12 Data Set
10.4.12 PROSTATE MRI Data Set
10.4.13 Prostate MR Image Database
10.4.14 BraTS Data Set
10.4.15 Brain-Tumor-Progression Data Set
10.5 Performance of Deep Learning-Based Models for Cancer Detection and Classification
10.5.1 Confusion Matrix
10.5.2 Classification Accuracy
10.5.3 Precision
10.5.4 Recall
10.5.5 F1 Score
10.5.6 Sensitivity and Specificity
10.6 Conclusion
Notes
References
Chapter 11 Enhancing Deep Learning-Based Organ Segmentation for Diagnostic Support Systems on Chest X-rays
11.1 Introduction
11.2 Materials and Methods
11.2.1 Convolutional Neural Networks
11.2.2 Chest X-Ray Database
11.3 Experimental Results
11.4 Discussion
11.5 Conclusion
References
Chapter 12 Deep Learning in Healthcare: A Bibliometric Analysis
12.1 Introduction
12.2 Methods
12.3 Results
12.3.1 Results of the Bibliometric Analysis of Authors
12.3.2 Results of the Bibliometric Analysis of Keywords
12.4 Discussion
12.5 Conclusion
Acknowledgment
References
Chapter 13 The Effects of Image Pre-processing in Detection of Brain Tumors Using Faster R-CNN on Magnetic Resonance Scans
13.1 Introduction
13.2 Materials and Methods
13.2.1 Brain MR Data Set
13.2.2 Median Filter
13.2.3 Second-Order Volterra Filters
13.2.4 Faster Regional-Convolutional Neural Network
13.3 Experimental Results
13.4 Discussion and Conclusions
References
Chapter 14 Evaluation of Deep Neural Network and Ensemble Machine Learning Methods for Cesarean Data Classification
14.1 Introduction
14.2 Materials and Methods
14.2.1 Data Set
14.2.2 Machine Learning Methods
14.2.3 Training of DNN and Ensemble Models
14.3 Results
14.4 Conclusion
References
Chapter 15 Transfer Learning for Classification of Brain Tumor
15.1 Introduction
15.2 Background
15.3 Methodology and Proposed Models
15.4 Conclusion
References
Chapter 16 Comparative Study on Analysis of Medical Images Using Deep Learning Techniques
16.1 Introduction
16.2 Machine Learning, Artificial Neural Networks, Deep Learning
16.3 Machine Learning and Deep Learning in Medical Imaging
16.3.1 Grouping
16.3.2 ID: Organ and Country Organ
16.3.3 Part Section
16.3.4 Selection
16.4 Medical Image Analysis using Deep Learning
16.4.1 Grouping of Goods
16.4.2 Restriction
16.4.3 Location
16.4.4 Part Section
16.5 Conclusion
References
Index