Deep Learning in Medical Image Processing and Analysis

Cover
Contents
About the editors
1 Diagnosing and imaging in oral pathology by use of artificial intelligence and deep learning
1.1 Introduction
1.1.1 Application of Artificial Intelligence in the Field of Oral Pathology
1.1.2 AI as oral cancer prognostic model
1.1.3 AI for oral cancer screening, identification, and classification
1.1.4 Oral cancer and deep ML
1.1.5 AI in predicting the occurrence of oral cancer
1.1.6 AI for oral tissue diagnostics
1.1.7 AI for OMICS in oral cancer
1.1.8 AI accuracy for histopathologic images
1.1.9 Mobile mouth screening
1.1.10 Deep learning in oral pathology image analysis
1.1.11 Future prospects and challenges
1.2 Conclusion
References
2 Oral implantology with artificial intelligence and applications of image analysis by deep learning
2.1 Introduction
2.2 Clinical application of AI’s machine learning algorithms in dental practice
2.2.1 Applications in orthodontics
2.2.2 Applications in periodontics
2.2.3 Applications in oral medicine and maxillofacial surgery
2.2.4 Applications in forensic dentistry
2.2.5 Applications in cariology
2.2.6 Applications in endodontics
2.2.7 Applications in prosthetics, conservative dentistry, and implantology
2.3 Role of AI in implant dentistry
2.3.1 Use of AI in radiological image analysis for implant placement
2.3.2 Deep learning in implant classification
2.3.3 AI techniques to detect implant bone level and marginal bone loss around implants
2.3.4 Comparison of the accuracy performance of dental professionals in classification with and without the assistance of the DL
2.3.5 AI in fractured dental implant detection
2.4 Software initiatives for dental implant
2.5 AI models and implant success predictions
2.6 Discussion
2.7 Final considerations
References
3 Review of machine learning algorithms for breast and lung cancer detection
3.1 Introduction
3.2 Literature review
3.3 Review and discussions
3.4 Proposed methodology
3.5 Conclusion
References
4 Deep learning for streamlining medical image processing
4.1 Introduction
4.2 Deep learning: a general idea
4.3 Deep learning models in medicine
4.3.1 Convolutional neural networks
4.3.2 Recurrent neural networks
4.3.3 Auto-encoders (AE)
4.4 Deep learning for medical image processing: overview
4.5 Literature review
4.6 Medical imaging techniques and their use cases
4.6.1 X-Ray image
4.6.2 Computerized tomography
4.6.3 Mammography
4.6.4 Histopathology
4.6.5 Endoscopy
4.6.6 Magnetic resonance imaging
4.6.7 Bio-signals
4.7 Application of deep learning in medical image processing and analysis
4.7.1 Segmentation
4.7.2 Classification
4.7.3 Detection
4.7.4 Deep learning-based tracking
4.7.5 Using deep learning for image reconstruction
4.8 Training testing validation of outcomes
4.9 Challenges in deploying deep learning-based solutions
4.10 Conclusion
References
5 Comparative analysis of lumpy skin disease detection using deep learning models
5.1 Introduction
5.1.1 Health issues of cattle
5.2 Related works
5.2.1 LSD diagnosis and prognosis
5.2.2 Other skin disease detection technique in cows
5.3 Proposed model
5.3.1 Data collection
5.3.2 Deep learning models
5.4 Experimental results and discussions
5.4.1 MLP model
5.4.2 CNN model
5.4.3 CNN+LSTM model
5.4.4 CNN+GRU model
5.4.5 Hyperparameters
5.4.6 Performance evaluation
5.5 Conclusion
References
6 Can AI-powered imaging be a replacement for radiologists?
6.1 Artificial Intelligence (AI) and its present footprints in radiology
6.2 Brief history of AI in radiology
6.3 AI aided medical imaging
6.4 AI imaging pathway
6.5 Prediction of disease
6.5.1 Progression without deep learning
6.5.2 Progress prediction with deep learning
6.6 Recent implementation of AI in radiology
6.6.1 Imaging of the thorax
AI model can aid in chest X-ray collapsed lung detection
6.6.2 Pelvic and abdominal imaging
A three-dimensional pelvic model utilizing artificial intelligence technologies for preoperative MRI simulation of rectal cancer
6.6.3 Colonoscopy
Colonoscopy using artificial intelligence assistance: a survey
6.6.4 Brain scanning
Current neuroimaging applications in the age of AI
6.6.5 Mammography
Thermal imaging and AI technology
6.7 How does AI help in the automated localization and segmentation of tumors?
6.7.1 Multi-parametric MR rectal cancer segmentation
6.7.2 Automated tumor characterization
6.8 The Felix Project
6.9 Challenges faced due to AI technology
6.10 Solutions to improve the technology
6.11 Conclusion
References
7 Healthcare multimedia data analysis algorithms tools and techniques
7.1 Introduction
7.1.1 Techniques for summarizing media data
7.1.2 Techniques for filtering out media data
7.1.3 Techniques for media description categorization—classes
7.2 Literature survey
7.3 Methodology
7.3.1 Techniques for data summarization
7.3.2 Merging and filtering method
7.3.3 Evaluating approaches
7.4 Sample illustration: case study
7.5 Applications
7.6 Conclusion
References
8 Empirical mode fusion of MRI-PET images using deep convolutional neural networks
8.1 Introduction
8.2 Preliminaries
8.2.1 Positron emission tomography resolution enhancement neural network (PET-RENN)
8.3 Multichannel bidimensional EMD through a morphological filter
8.4 Proposed method
8.4.1 EMD
8.4.2 Fusion rule
8.5 Experiments and results
8.5.1 Objective metrics
8.5.2 Selected specifications
8.6 Conclusion
References
9 A convolutional neural network for scoring of sleep stages from raw single-channel EEG signals
9.1 Introduction
9.2 Background study
9.3 Methodology
9.3.1 Sleep dataset
9.3.2 Preprocessing
9.3.3 CNN classifier architecture
9.3.4 Optimization
9.4 Criteria for evaluation
9.5 Training algorithm
9.5.1 Pre-training
9.5.2 Supervised fine-tuning
9.5.3 Regularization
9.6 Results
9.7 Discussion
9.7.1 Major findings
9.7.2 The problem of class imbalance
9.7.3 Comparison
References
10 Fundamentals, limitations, and the prospects of deep learning for biomedical image analysis
10.1 Introduction
10.2 Demystifying DL
10.3 Current trends in intelligent disease detection systems
10.3.1 Overview
10.3.2 Radiology
10.3.3 Ophthalmology
10.3.4 Dermatology
10.4 Challenges and limitations in building biomedical image processing systems
10.5 Patient benefits
10.6 Conclusions
References
11 Impact of machine learning and deep learning in medical image analysis
11.1 Introduction
11.2 Overview of machine learning methods
11.2.1 Supervised learning
11.2.2 Unsupervised learning
11.2.3 Reinforcement learning
11.3 Neural networks
11.3.1 Convolutional neural network
11.4 Why deep learning over machine learning
11.5 Deep learning applications in medical imaging
11.5.1 Histopathology
11.5.2 Computerized tomography
11.5.3 Mammograph
11.5.4 X-rays
11.6 Conclusion
Conflict of interest
References
12 Systemic review of deep learning techniques for high-dimensional medical image fusion
12.1 Introduction
12.2 Basics of image fusion
12.2.1 Pixel-level medical image fusion
12.2.2 Transform-level medical image fusion
12.2.3 Multi-modal fusion in medical imaging
12.3 Deep learning methods
12.3.1 Image fusion based on CNNs
12.3.2 Image fusion by morphological component analysis
12.3.3 Image fusion by guided filtering
12.3.4 Image fusion based on generative adversarial network (GAN)
12.3.5 Image fusion based on autoencoders
12.4 Optimization methods
12.4.1 Evaluation
12.5 Conclusion
References
13 Qualitative perception of a deep learning model in connection with malaria disease classification
13.1 Image classification
13.1.1 Deep learning
13.2 Layers of convolution layer
13.2.1 Convolution neural network
13.2.2 Pointwise and depthwise convolution
13.3 Proposed model
13.4 Implementation
13.5 Result
13.6 Conclusion
References
14 Analysis of preperimetric glaucoma using a deep learning classifier and CNN layer-automated perimetry
14.1 Introduction
14.2 Literature survey
14.3 Methodology
14.3.1 Procedure for eye detection
14.3.2 Deep CNN architecture
14.4 Experiment analysis and discussion
14.4.1 Pre-processing
14.4.2 Performance analysis
14.4.3 CNN layer split-up analysis
14.5 Conclusion
References
15 Deep learning applications in ophthalmology—computer-aided diagnosis
15.1 Introduction
15.2 Ophthalmology
15.2.1 Diabetic retinopathy
15.2.2 Age-related macular degeneration
15.2.3 Glaucoma
15.2.4 Cataract
15.3 Neuro-ophthalmology
15.3.1 Papilledema
15.3.2 Alzheimer’s disease
15.4 Systemic diseases
15.4.1 Chronic kidney disease
15.4.2 Cardiovascular diseases
15.5 Challenges and opportunities
15.6 Future trends
15.6.1 Smartphone image capture
15.7 Multi-disease detection using a single retinal fundus image
15.8 Conclusion
15.9 Abbreviations used
References
16 Brain tumor analyses adopting a deep learning classifier based on glioma, meningioma, and pituitary parameters
16.1 Introduction
16.2 Literature survey
16.3 Methodology
16.3.1 Procedure for brain tumor detection
16.3.2 Deep CNN (DCNN) architecture
16.4 Experiment analysis and discussion
16.4.1 Preprocessing
16.4.2 Performance analysis
16.4.3 Brain tumor deduction
16.4.4 CNN layer split-up analysis
16.5 Conclusion
References
17 Deep learning method on X-ray image super-resolution based on residual mode encoder–decoder network
17.1 Introduction
17.2 Preliminaries
17.2.1 Encoder–decoder residual network
17.3 Coarse-to-fine approach
17.4 Residual in residual block
17.5 Proposed method
17.5.1 EDRN
17.6 Experiments and results
17.6.1 Datasets and metrics
17.6.2 Training settings
17.6.3 Decoder–encoder architecture
17.6.4 Coarse-to-fine approach
17.6.5 Investigation of batch normalization
17.6.6 Results for classic single image X-ray super-resolution
17.7 Conclusion
References
18 Melanoma skin cancer analysis using convolutional neural networks-based deep learning classification
18.1 Introduction
18.2 Literature survey
18.3 Methodology
18.3.1 MobileNetv2
18.3.2 Inception v3
18.4 Results
18.4.1 Data pre-processing
18.4.2 Performance analysis
18.4.3 Statistical analysis
18.5 Conclusion
References
19 Deep learning applications in ophthalmology and computer-aided diagnostics
19.1 Introduction
19.1.1 Motivation
19.2 Technical aspects of deep learning
19.3 Anatomy of the human eye
19.4 Some of the most common eye diseases
19.4.1 Diabetic retinopathy (DR)
19.4.2 Age-related macular degeneration (AMD or ARMD)
19.4.3 Glaucoma
19.4.4 Cataract
19.4.5 Macular edema
19.4.6 Choroidal neovascularization
19.5 Deep learning in eye disease classification
19.5.1 Diabetic retinopathy
19.5.2 Glaucoma
19.5.3 Age-related macular degeneration
19.5.4 Cataracts and other eye-related diseases
19.6 Challenges and limitations in the application of DL in ophthalmology
19.6.1 Challenges in the practical implementation of DL ophthalmology
19.6.2 Technology-related challenges
19.6.3 Social and cultural challenges for DL in the eyecare
19.6.4 Limitations
19.7 Future directions
19.8 Conclusion
References
20 Deep learning for biomedical image analysis in
place of fundamentals, limitations, and prospects
of deep learning for biomedical image analysis
20.1 Introduction
20.2 Biomedical imaging
20.2.1 Computed tomography
20.2.2 Magnetic resonance imaging
20.2.3 Positron emission tomography
20.2.4 Ultrasound
20.2.5 X-ray imaging
20.3 Deep learning
20.3.1 Artificial neural network
20.4 DL models with various architectures
20.4.1 Deep neural network
20.4.2 Convolutional neural network
20.4.3 Recurrent neural network
20.4.4 Deep convolutional extreme learning machine
20.4.5 Deep Boltzmann machine
20.4.6 Deep autoencoder
20.5 DL in medical imaging
20.5.1 Image categorization
20.5.2 Image classification
20.5.3 Detection
20.5.4 Segmentation
20.5.5 Data mining
20.5.6 Registration
20.5.7 Other aspects of DL in medical imaging
20.5.8 Image enhancement
20.5.9 Integration of image data into reports
20.6 Summary of review
20.7 Challenges of DL in medical imaging
20.7.1 Large amount of training dataset
20.7.2 Legal and data privacy issues
20.7.3 Standards for datasets and interoperability
20.7.4 Black box problem
20.7.5 Noise labeling
20.7.6 Images of abnormal classes
20.8 The future of DL in biomedical image processing
20.9 Conclusion
References
Index
Back Cover