Deep Learning Models for Medical Imaging
Deep Learning Models for Medical Imaging explains the concepts of Deep Learning (DL) and its importance in medical imaging and/or healthcare using two different case studies: a) cytology image analysis and b) coronavirus (COVID-19) prediction, screening, and decision-making, using publicly available...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | eBook |
| Language: | English |
| Published: |
Chantilly
Elsevier Science & Technology
2021
Academic Press |
| Edition: | 1 |
| Series: | Primers in Biomedical Imaging Devices and Systems |
| Subjects: | |
| ISBN: | 9780128235041, 0128235047 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- Front Cover -- Deep Learning Models for Medical Imaging -- Copyright -- Contents -- List of figures -- List of tables -- Authors -- KC Santosh -- Nibaran Das -- Swarnendu Ghosh -- Foreword -- Preface -- Acronyms -- 1 Introduction -- 1.1 Background -- 1.2 Machine learning and its types -- 1.3 Evolution of machine learning -- 1.3.1 Rule-based learning -- 1.3.2 Feature-based learning -- 1.3.3 Representation learning -- 1.4 Basics to deep learning -- 1.4.1 The rise of cybernetics -- 1.4.2 The connectionist movement -- 1.4.3 The onset of deep learning -- 1.4.4 Motivation: deep learning -- 1.5 Importance of deep learning -- 1.6 Deep learning in medical imaging: a review -- 1.6.1 Medical imaging scope -- 1.6.2 Medical imaging data -- 1.6.3 Applications: deep learning in medical imaging -- 1.7 Scope of the book -- References -- 2 Deep learning: a review -- 2.1 Background -- 2.2 Artificial neural networks -- 2.2.1 The neuron -- 2.2.2 Activation functions -- 2.2.3 Multilayer feed forward neural network -- 2.2.4 Training neural networks by back-propagation -- 2.2.5 Optimization -- 2.2.5.1 Objective functions -- Mean squared error -- Cross-entropy measures -- 2.2.5.2 Optimization techniques -- Stochastic gradient descent -- Momentum -- Adaptive learning rates -- 2.2.6 Regularization -- 2.3 Convolutional neural networks -- 2.3.1 Feature extraction using convolutions -- 2.3.2 Subsampling -- 2.3.3 Effect of nonlinearity on activation maps -- 2.3.4 Layer design -- 2.3.5 Output layer -- 2.4 Encoder-decoder architecture -- 2.4.1 Unsupervised learning in CNNs -- 2.4.2 Image-to-image translation -- 2.4.3 Localization -- 2.4.4 Multiscale feature propagation -- References -- 3 Deep learning models -- 3.1 Deep learning models -- 3.1.1 Learning different objectives -- 3.1.2 Network structure for CNNs -- 3.1.3 Types of models based on learning strategies
- 3.2 Elements in deep learning pipeline -- 3.2.1 Data preprocessing -- 3.2.2 Model selection -- 3.2.3 Model validation and hyperparameter tuning -- 3.3 Evolution of deep learning models and applications -- 3.3.1 Classification -- 3.3.2 Localization -- 3.3.3 Segmentation -- References -- 4 Cytology image analysis -- 4.1 Background -- 4.2 Cytology: a brief overview -- 4.3 Types of cytology -- 4.4 Cytology slide preparation -- 4.4.1 Aspiration cytology -- 4.4.2 Exfoliative cytology -- 4.4.3 Abrasive cytology -- 4.4.4 Specimen collection -- 4.4.5 Slide preparation -- 4.4.6 Fixation techniques and staining protocol -- 4.5 Cytological process and digitization -- 4.6 Cervical cell cytology -- 4.6.1 Modalities of cervical specimen collection -- 4.6.2 Characteristics of cytomorphology of malignant cells -- 4.7 Experiments -- 4.7.1 Dataset -- 4.7.2 Experimental setup and protocols -- 4.7.2.1 Transfer learning: a quick overview -- 4.7.3 Results and discussion -- 4.7.3.1 Results with or without using transfer learning -- 4.7.3.2 Results with data augmentation -- 4.7.3.3 Results using ensemble of classifiers -- 4.7.4 Summary -- References -- 5 COVID-19: prediction, screening, and decision-making -- 5.1 Background -- 5.2 Predictive modeling and infectious disease outbreaks -- 5.3 Need of medical imaging tools for COVID-19 outbreak screening -- 5.4 Deep neural networks for COVID-19 screening -- 5.4.1 Truncated Inception Net: COVID-19 outbreak screening using chest X-rays [7] -- 5.4.2 Shallow CNN for COVID-19 outbreak screening using chest X-rays [2] -- 5.4.3 DNN to detect COVID-19: one architecture for both chest CT and X-ray images [3] -- 5.5 Discussion: how big data is big? -- References -- Index -- Back Cover