Within this first volume dealing with lung and kidney cancer, the editors and authors will detail the latest research related to the application of AI to cancer diagnosis and prognosis and summarize its advantages. It's the editors and authors intention to explore how AI assists in these activities, specifically with regard to its unprecedented accuracy, which is even higher than that of general statistical applications in oncology. Ways will also be demonstrated as to how these methods in AI are advancing the field.
There have been thousands of papers written between 1995 and 2019 related to AI for cancer diagnosis and prognosis. However, to this date (and unknown to the Editors) there has not yet been published a comprehensive overview of the latest findings pertaining to these AI technologies, within a single book project(s). Therefore, the purpose of this three volume work and particularly for this first volume dealing with lung and kidney cancer, is to present a compendium of these findings related to these two pervasive cancers. Within this coverage it's our hope that scientists, researchers and clinicians can successfully incorporate these techniques into other significant cancers such as pancreatic, esophageal leukemia, melanoma, etc.
Key Features:
- This work will contain a comprehensive overview of the latest techniques in Artificial Intelligence (AI) related to lung and kidney cancers.
- All chapter authors and contributors will be world-class researchers in various aspects of AI and appropriate subsets such as machine learning (ML), deep learning (DL) and neural networks.
- The fusion of 'Big Data' and 'AI' will be incorporated where appropriate.
- Multimodality imaging will be included within specific chapters.
- Extensive references will be included at the end of each chapter to enhance further study.
Author(s): Ayman El-Baz, Jasjit S. Suri
Series: IPEM–IOP Series in Physics and Engineering in Medicine and Biology
Publisher: IOP Publishing
Year: 2022
Language: English
Pages: 249
City: Bristol
PRELIMS.pdf
Preface
Acknowledgements
Editor biographies
Ayman El-Baz
Jasjit S Suri
List of contributors
CH001.pdf
Chapter 1 American Joint Committee on Cancer staging of lung and renal cancers using a recurrent deep neural network model
1.1 Introduction
1.2 Background
1.2.1 Lung cancer
1.2.2 Renal cancer
1.2.3 Research scope
1.3 Methodology
1.3.1 AJCC staging
1.3.2 Database
1.3.3 The deep learning model
1.4 The experiment
1.5 Results and discussion
1.6 Conclusions
References
CH002.pdf
Chapter 2 Neural-ensemble-based detection: a modern way to diagnose lung cancer
2.1 Introduction
2.1.1 Lung cancer epidemiology
2.1.2 Signs and symptoms of lung cancer
2.1.3 Staging of lung cancer
2.1.4 Classification of lung cancer
2.2 Different methods of lung cancer detection
2.2.1 Invasive methods
2.2.2 Non-invasive methods
2.3 Neural-ensemble-based detection
2.4 Conclusions
References and further reading
CH003.pdf
Chapter 3 Computed tomography and magnetic resonance imaging machine learning applications for renal cell carcinoma
3.1 Background
3.2 Applications
3.2.1 Malignant versus benign discrimination
3.2.2 Malignancy subtyping
3.2.3 Biologic aggressiveness
3.2.4 Correlation with overall and progression-free survival under treatment
3.2.5 Prediction of perioperative complications
3.3 Conclusions
References
CH004.pdf
Chapter 4 Pulmonary nodule-based feature learning for automated lung tumor grading using convolutional neural networks
4.1 Introduction
4.2 Literature review
4.2.1 Preprocessing
4.2.2 Candidate nodule segmentation
4.2.3 Feature extraction and classification
4.3 Methodology
4.3.1 Data acquisition
4.3.2 Preprocessing
4.3.3 NROI segmentation
4.3.4 GAN
4.3.5 Feature extraction
4.3.6 Classification
4.4 Results and discussion
4.5 Conclusions
References
CH005.pdf
Chapter 5 Detection of lung contours using closed principal curves and machine learning
5.1 Introduction
5.2 Materials and methods
5.2.1 Principal curve
5.2.2 Machine learning
5.2.3 Proposed algorithm
5.2.4 Quantitative evaluation
5.3 Results and discussion
5.3.1 Detecting contours in the private dataset using different learning rates
5.3.2 Detecting contours in the private dataset using different numbers of neurons in the hidden layer
5.3.3 Detecting contours in the private dataset using different numbers of epochs
5.3.4 Detecting contours in the private dataset using different algorithms
5.3.5 Detecting contours in the public LIDC–IDRI dataset using different algorithms
5.4 Conclusions
Acknowledgments
References
CH006.pdf
Chapter 6 Bytes, pixels, and bases: machine learning in imaging–omics for renal cell carcinoma
6.1 Introduction
6.1.1 The convergence of computers and cancer care
6.2 Imaging in renal cell carcinoma
6.2.1 Radiology
6.2.2 Pathology
6.3 Omics in renal cell carcinoma
6.3.1 Multiomics
6.4 Imaging–omics for kidney carcinoma
6.4.1 Radiomics
6.4.2 Pathomics
6.5 Opportunities and obstacles
6.5.1 Data
6.5.2 Interpretability
6.5.3 Privacy
6.5.4 Adversarial attacks
6.5.5 Regulatory roadblocks
6.6 Future directions
6.7 Conclusions
References
CH007.pdf
Chapter 7 Detection, growth quantification, and malignancy prediction of pulmonary nodules using deep convolutional networks in follow-up CT scans
7.1 Introduction
7.2 Background
7.2.1 Nodule detection
7.2.2 Nodule quantification
7.2.3 Lung cancer prediction
7.3 Temporal lung nodule assessment
7.3.1 Preprocessing
7.3.2 Nodule detection
7.3.3 Nodule reidentification
7.3.4 Nodule growth quantification
7.3.5 Nodule malignancy classification
7.4 Data cohort
7.4.1 Scanners and protocols
7.4.2 Data
7.5 Results
7.5.1 Nodule detection
7.5.2 Nodule reidentification
7.5.3 Nodule growth quantification
7.5.4 Nodule malignancy classification
7.6 Discussion
7.7 Conclusions
References and further reading
CH008.pdf
Chapter 8 Training a deep multiview model using small samples of medical data
8.1 Introduction
8.2 Related work
8.2.1 Cox proportional hazard model
8.2.2 Deep survival models
8.3 Methodology
8.3.1 Training the deep multiview model on small numbers of data samples
8.3.2 Training the network using a divide-and-conquer strategy
8.3.3 Training the model as a multitask model (MM)
8.4 Experiments and discussion
8.4.1 Data set descriptions
8.4.2 Data preprocessing
8.4.3 Experimental setup
8.4.4 Results
8.4.5 Discussion
8.5 Conclusions
References
CH009.pdf
Chapter 9 Overview of deep learning for lung cancer diagnosis
9.1 Introduction
9.2 Deep learning
9.2.1 Convolutional neural networks
9.2.2 Transfer learning models
9.2.3 The U-Net
9.3 Evaluation criteria
9.3.1 Evaluation metrics used in classification applications
9.3.2 Evaluation metrics used in segmentation applications
9.4 Datasets
9.4.1 The LIDC–IDRI data set
9.4.2 The LungCT-Diagnosis data set
9.4.3 The NSCLC-Radiomics data set
9.5 Overview of recent research
9.6 Discussion
9.7 Conclusions
References
CH010.pdf
Chapter 10 Artificial intelligence for cancer diagnosis
10.1 Introduction
10.2 Background of cancer
10.3 The basics of artificial intelligence
10.4 AI impacts on cancer-based clinical analysis
10.5 Visualization tools for AI-assisted cancer recognition systems
10.6 Multi-platform deployment for cancer prognosis systems
10.7 Case studies of cancer recognition systems that use artificial intelligence techniques
10.8 Conclusions
References and further reading
CH011.pdf
Chapter 11 Lung cancer diagnosis using 3D-CNN and spherical harmonics expansions
11.1 Introduction
11.2 Methodology
11.2.1 Feature extraction utilizing convolutional neural networks
11.2.2 Explanation of feature extraction utilizing spherical harmonics
11.3 Results
11.3.1 Experimental setup
11.3.2 Experimental evaluation
11.4 Conclusions
References
