State of the Art in Neural Networks and Their Applications: Volume 1

✍ Scribed by Ayman S. El-Baz, Jasjit S. Suri

• Publisher: Academic Press
• Year: 2021
• Tongue: English
• Leaves: 310
• Edition: 1
• Category: Library

✦ Synopsis

State of the Art in Neural Networks and Their Applications presents the latest advances in artificial neural networks and their applications across a wide range of clinical diagnoses. Advances in the role of machine learning, artificial intelligence, deep learning, cognitive image processing and suitable data analytics useful for clinical diagnosis and research applications are covered, including relevant case studies. The application of Neural Network, Artificial Intelligence, and Machine Learning methods in biomedical image analysis have resulted in the development of computer-aided diagnostic (CAD) systems that aim towards the automatic early detection of several severe diseases.

State of the Art in Neural Networks and Their Applications is presented in two volumes. Volume 1 covers the state-of-the-art deep learning approaches for the detection of renal, retinal, breast, skin, and dental abnormalities and more.

✦ Table of Contents

Title-page_2021_State-of-the-Art-in-Neural-Networks-and-their-Applications
State of the Art in Neural Networks and Their Applications
Copyright_2021_State-of-the-Art-in-Neural-Networks-and-their-Applications
Copyright
Dedication_2021_State-of-the-Art-in-Neural-Networks-and-their-Applications
Dedication
Contents_2021_State-of-the-Art-in-Neural-Networks-and-their-Applications
Contents
List-of-Contributo_2021_State-of-the-Art-in-Neural-Networks-and-their-Applic
List of Contributors
Biographies_2021_State-of-the-Art-in-Neural-Networks-and-their-Applications
Biographies
Acknowledgments_2021_State-of-the-Art-in-Neural-Networks-and-their-Applicati
Acknowledgments
Chapter-1---Computer-aided-detection-of-abno_2021_State-of-the-Art-in-Neural
1 Computer-aided detection of abnormality in mammography using deep object detectors
1.1 Introduction
1.2 Literature review
1.3 Methodology
1.3.1 Architectures of deep convolutional neural networks and deep object detectors
1.3.2 Abnormality detection with faster R-convolutional neural networks
1.3.3 Abnormality detection with YOLO
1.4 Experimental results
1.4.1 Data preparation
1.4.2 Abnormality detection with faster R-convolutional neural networks
1.4.3 Abnormality detection with YOLO
1.4.4 Results comparison
1.5 Discussions
1.6 Conclusion
References
Chapter-2---Detection-of-retinal-abnormaliti_2021_State-of-the-Art-in-Neural
2 Detection of retinal abnormalities in fundus image using CNN deep learning networks
2.1 Introduction
2.2 Earlier screening and diagnosis of ocular diseases with CNN deep learning networks
2.2.1 Glaucoma
2.2.1.1 Methods and materials
2.2.1.2 Deep learning neural-network architectures for glaucoma screening and diagnosis
2.2.1.3 Application and evaluation on earlier glaucoma screening and diagnosis—classification
2.2.1.3.1 Fundus image glaucoma classification
2.2.1.3.2 Optical coherence tomography image glaucoma classification
2.2.1.4 Datasets used in glaucoma diagnosis
2.2.2 Age-related macular degeneration
2.2.2.1 Methods and materials
2.2.2.2 Deep learning–based methods for age-related macular degeneration detection and grading
2.2.3 Diabetic retinopathy
2.2.3.1 Methods and materials
2.2.3.2 Deep learning–based methods for diabetic retinopathy detection and grading
2.2.3.3 Dataset used diabetic retinopathy diagnosis
2.2.4 Cataract
2.2.4.1 Methods and materials
2.2.4.2 Deep learning–based methods for cataract detection and grading
2.3 Deep learning–based smartphone for detection of retinal abnormalities
2.3.1 Smartphone-captured fundus image evaluation
2.3.2 Deep learning–based method of ocular pathology detection from smartphone-captured fundus image
2.4 Discussion
2.5 Conclusion
References
Chapter-3---A-survey-of-deep-learning-based-_2021_State-of-the-Art-in-Neural
3 A survey of deep learning-based methods for cryo-electron tomography data analysis
3.1 Introduction
3.2 Deep learning-based methods
3.2.1 Detection and segmentation
3.2.2 Classification
3.2.3 Others
3.3 Conclusion
References
Chapter-4---Detection--segmentation--and-numbering-_2021_State-of-the-Art-in
4 Detection, segmentation, and numbering of teeth in dental panoramic images with mask regions with convolutional neural ne...
4.1 Introduction
4.2 Related work
4.3 Fédération Dentaire Internationale tooth numbering system
4.4 The method
4.4.1 Implementation details
4.4.1.1 Tooth numbering
4.5 Experimental analysis
4.5.1 Dataset
4.5.2 Evaluation
4.5.3 Results
4.6 Discussion and conclusions
References
Chapter-5---Accurate-identification-of-renal-tr_2021_State-of-the-Art-in-Neu
5 Accurate identification of renal transplant rejection: convolutional neural networks and diffusion MRI
5.1 Introduction
5.2 Methods
5.2.1 Kidney segmentation
5.2.2 Feature extraction
5.2.3 Renal transplant classification using deep convolutional neural network
5.3 Experimental results
5.4 Conclusion
Acknowledgments
References
Chapter-6---Applications-of-the-ESPNet-_2021_State-of-the-Art-in-Neural-Netw
6 Applications of the ESPNet architecture in medical imaging
6.1 Introduction
6.2 Background
6.2.1 Standard convolution
6.2.2 Dilated convolution
6.3 The ESPNet architecture
6.3.1 Efficient spatial pyramid unit
6.3.1.1 Hierarchical feature fusion for degridding in the efficient spatial pyramid unit
6.3.2 Segmentation architecture
6.4 Experimental results
6.4.1 Breast biopsy whole slide image dataset
6.4.1.1 Dataset
6.4.1.2 Training
6.4.1.3 Segmentation results
6.4.1.4 Skip connections
6.4.1.5 Pyramidal spatial pooling as a decoding unit
6.4.1.6 Comparison with state-of-the-art methods
6.4.1.7 Tissue-level segmentation masks for computer-aided diagnosis
6.4.2 Brain tumor segmentation
6.4.2.1 Dataset
6.4.2.2 Training
6.4.2.3 Results
6.4.3 Other applications
6.5 Conclusion
Acknowledgment
References
Chapter-7---Achievements-of-neural-netwo_2021_State-of-the-Art-in-Neural-Net
7 Achievements of neural network in skin lesions classification
7.1 Introduction
7.2 Literature review
7.3 Background
7.4 Dataset
7.5 Methodology
7.6 Results and discussion
7.7 Conclusion
References
Chapter-8---A-computer-aided-diagnosis-system-f_2021_State-of-the-Art-in-Neu
8 A computer-aided diagnosis system for breast cancer molecular subtype prediction in mammographic images
8.1 Introduction
8.2 Background
8.2.1 Breast cancer detection
8.2.2 Breast tumor segmentation
8.2.3 Shape classification
8.3 Datasets
8.4 Methodology
8.4.1 Modified Faster R-CNN for breast tumor detection
8.4.2 Breast tumor segmentation using conditional generative adversarial network
8.4.3 Shape descriptor using convolutional neural network
8.4.4 Breast cancer molecular subtypes correlation to the tumor shape
8.5 Conclusion
References
Chapter-9---Computer-aided-diagnos_2021_State-of-the-Art-in-Neural-Networks-
9 Computer-aided diagnosis of renal masses
9.1 Introduction
9.2 Segmentation of kidneys
9.2.1 Convolutional neural network
9.2.1.1 Convolutional layer
9.2.1.2 Detection layer
9.2.1.3 Pooling layer
9.2.2 U-Net
9.2.3 Performance evaluation of the algorithm
9.2.4 Deep learning–based methods for automated segmentation of kidney
9.3 Kidney tumor localization
9.4 Differentiation of malignant versus benign renal masses
9.5 Future perspectives
References
Chapter-10---Early-identification-of-acute-re_2021_State-of-the-Art-in-Neura
10 Early identification of acute rejection for renal allografts: a machine learning approach
Acknowledgment
10.1 Introduction
10.2 Methods
10.2.1 Diffusion-weighted image markers
10.2.2 Clinical biomarkers
10.2.3 Integration process of clinical with imaging biomarkers
10.3 Experimental results
10.4 Conclusion
References
Chapter-11---Deep-learning-for-computer-ai_2021_State-of-the-Art-in-Neural-N
11 Deep learning for computer-aided diagnosis in ophthalmology: a review
11.1 Introduction
11.1.1 The burden of eye disease
11.1.2 Imaging and image analysis
11.1.3 Deep learning: an emerging state-of-the-art
11.2 Deep learning: the methods
11.2.1 Reference standards
11.2.2 Preprocessing and augmentation
11.2.3 Architectures, transfer learning, and ensembling
11.2.4 Loss functions and performance metrics
11.3 Limitations of the state-of-the-art
11.3.1 Trustworthiness and transparency
11.3.2 Uncertainty estimation
11.3.3 Explainability and interpretability
11.4 Beyond convolutional neural networks
11.4.1 Generative adversarial networks
11.4.2 Capsule networks
11.5 Conclusion
References
Chapter-12---Deep-learning-for-ophthalmol_2021_State-of-the-Art-in-Neural-Ne
12 Deep learning for ophthalmology using optical coherence tomography
12.1 Introduction
12.2 Optical coherence tomography
12.2.1 Variations of optical coherence tomography systems
12.2.1.1 Time-domain optical coherence tomography
12.2.1.2 Spectral domain optical coherence tomography
12.2.1.3 Polarization-sensitive optical coherence tomography
12.2.1.4 Swept-source optical coherence tomography
12.2.2 Optical coherence tomography datasets
12.2.3 Advantages and disadvantages of optical coherence tomography imaging
12.3 Retinal biomarkers and diseases
12.3.1 Important retinal diseases
12.3.1.1 Diabetic retinopathy and diabetic macular edema
12.3.1.2 Glaucoma
12.3.1.3 Age-related macular degeneration
12.3.2 Biomarker use in disease analysis
12.4 Traditional approaches for ophthalmic diagnosis
12.4.1 Image-processing fundamentals
12.4.2 Feature extraction fundamentals
12.4.3 Traditional classifiers
12.4.4 Applications
12.4.4.1 Denoising
12.4.4.2 Segmentation
12.4.4.3 Classification
12.5 Deep learning approaches to optical coherence tomography analysis
12.5.1 Convolutional neural network applications
12.5.2 Autoencoder applications
12.5.3 Generative adversarial network applications
12.6 Final thoughts
Acknowledgment
Conflict of interest
References
Further reading
Chapter-13---Generative-adversarial-n_2021_State-of-the-Art-in-Neural-Networ
13 Generative adversarial networks in medical imaging
13.1 Introduction
13.2 Applications in medical imaging
13.2.1 Localization and classification
13.2.2 CS-MRI reconstruction
13.2.3 Customized medical products
13.3 Conclusions
References
Chapter-14---Deep-learning-from-small-label_2021_State-of-the-Art-in-Neural-
14 Deep learning from small labeled datasets applied to medical image analysis
14.1 Introduction
14.2 Cross-modality deep learning
14.2.1 Feature adaptation-based cross-modality learning
14.2.2 Data augmentation-based cross-modality learning
14.3 Example of cross-domain adaptation-based segmentation: lung tumor segmentation from MRI
14.3.1 Study population and datasets
14.3.2 Method
14.4 Results
14.5 Future outlook and discussion
14.6 Conclusion
References
Index_2021_State-of-the-Art-in-Neural-Networks-and-their-Applications
Index