Radiomics and Radiogenomics in Neuro-Oncology

An Artificial Intelligence Paradigm – Volume 2: Genetics and Clinical Applications

1st Edition - October 15, 2024
Authors: Sanjay Saxena, Jasjit S. Suri
Language: English
Paperback ISBN:
9 7 8 - 0 - 4 4 3 - 1 8 5 0 9 - 0
eBook ISBN:
9 7 8 - 0 - 4 4 3 - 1 8 5 1 0 - 6

Radiomics and Radiogenomics in Neuro-Oncology: An Artificial Intelligence Paradigm—Volume 2: Genetics and Clinical Applications provides readers with a broad and detailed framew… Read more

Purchase options

Institutional subscription on ScienceDirect

Request a sales quote

Radiomics and Radiogenomics in Neuro-Oncology: An Artificial Intelligence Paradigm—Volume 2: Genetics and Clinical Applications provides readers with a broad and detailed framework for radiomics and radiogenomics (R-n-R) approaches with AI in neuro-oncology. It delves into the study of cancer biology and genomics, presenting methods and techniques for analyzing these elements. The book also highlights current solutions that R-n-R can offer for personalized patient treatments, as well as discusses the limitations and future prospects of AI technologies.
Volume 1: Radiogenomics Flow Using Artificial Intelligence covers the genomics and molecular study of brain cancer, medical imaging modalities and their analysis in neuro-oncology, and the development of prognostic and predictive models using radiomics.
Volume 2: Genetics and Clinical Applications extends the discussion to imaging signatures that correlate with molecular characteristics of brain cancer, clinical applications of R-n-R in neuro-oncology, and the use of Machine Learning and Deep Learning approaches for R-n-R in neuro-oncology.

Radiomics and Radiogenomics in Neuro-Oncology
Cover image
Title page
Table of Contents
Copyright
Contributors
About the editors
Preface
Acknowledgments
Section 1: Imaging signatures for brain cancer molecular characteristics
Chapter 1 Radiogenomics and genetic diversity of glioblastoma characterization
Abstract
Keywords
1.1 Introduction
1.2 An overview of glioblastoma multiforme (GBM)
1.3 Genetics of glioblastoma
1.4 Genetic defects in human cancer
1.5 Genetic diversity of glioblastoma multiforme
1.5.1 Classical subtype of GBM (21% of core samples)
1.5.2 Mesenchymal subtype of GBM (32% of core samples)
1.5.3 Proneural subtype of GBM (31% of core samples)
1.5.4 Neural subtype of GBM (16% of core samples)
1.5.5 Subtypes and clinical correlations
1.6 Genomics of glioblastoma
1.7 Radiogenomics of glioblastoma
1.8 Radiogenomic studies
1.8.1 Tumor core
1.8.2 Peritumoral edematous
1.8.3 Necrotic core
1.8.4 Diffusion restriction
1.8.5 Angiogenesis
1.8.6 Metabolic function
1.9 Considerations and challenges of radiogenomic studies
1.10 Future directions
1.11 Conclusion
References
Chapter 2 Machine and deep learning-based methods for genotype O(6)-methylguanine-DNA-methyltransferase status prediction
Abstract
Keywords
2.1 Introduction
2.2 Background studies
2.2.1 Radiogenomics and deep learning
2.2.2 Multimodal and multisequence imaging
2.2.3 Genetic and epigenetic biomarkers
2.3 Literature survey
2.4 Methods
2.4.1 Data collection
2.4.2 Preprocessing
2.4.3 Feature extraction and selection
2.4.4 Models
2.4.5 Performance evaluation
2.5 Challenges and future directions
2.6 Conclusion
References
Chapter 3 AI-based image signature for brain cancer molecular analysis
Abstract
Keywords
3.1 Introduction
3.2 Artificial intelligence and imaging technologies
3.3 Molecular imaging and AI scope
3.4 AI and imaging technology
3.5 Specialized microscopy and AI
3.6 Tumor characterization and radiomics approach
3.7 Brain tumor classification using artificial intelligence with a radiogenomics pipeline
3.8 Future and challenges of artificial intelligence in brain tumors
3.9 Conclusion
References
Chapter 4 Imaging signatures for different mutation estimation for brain cancer
Abstract
Keywords
4.1 Introduction
4.2 Different types of brain tumor
4.2.1 Gliomas
4.2.2 Meningiomas
4.2.3 Medulloblastomas
4.2.4 Schwannomas
4.2.5 Neurofibromas
4.2.6 Pituitary adenomas
4.2.7 Craniopharyngiomas
4.2.8 Primary CNS lymphoma (PCNSL)
4.2.9 Pediatric high-grade gliomas
4.3 Imaging modalities
4.3.1 Magnetic resonance imaging (MRI)
4.3.2 Positron emitted tomography (PET)
4.3.3 Computed tomography (CT)
4.4 ML models in imaging
4.4.1 ML models
4.4.2 Limitations
4.5 DL advances
4.5.1 Convolutional neural network (CNN)
4.5.2 Generative adversarial network (GAN)
4.5.3 U-net
4.5.4 Graph neural network (GNN)
4.5.5 Limitations
4.5.6 DL in gene mutation
4.6 Data handling
4.7 Integration with clinical workflow
4.8 Project description
4.8.1 Case study 1: Residual CNN for determination of IDH status in low-grade and high-grade gliomas from MR imaging
4.8.2 Case study 2: DL CNN accurately classify genetic mutations in gliomas
4.9 Challenges and limitations
4.10 Future directions
4.11 Conclusion
References
Chapter 5 Software solutions for managing radiomics and radiogenomics in neuro-oncology clinical settings
Abstract
Keywords
5.1 Introduction
5.2 Key components of radiomics and radiogenomics
5.2.1 Quantitative imaging features extraction methods
5.2.2 Convolutional neural networks
5.2.3 Features selection
5.2.4 Conventional machine learning models
5.2.5 Deep learning models
5.2.6 Advanced machine learning models
5.3 Key softwares
5.3.1 Scikit learn
5.3.2 PyTorch
5.3.3 TensorFlow
5.3.4 Weka
5.3.5 KNIME
5.3.6 Colab
5.3.7 Apache Mahout
5.3.8 Across.net
5.3.9 Shogun
5.3.10 Keras
5.3.11 Rapid miner
5.3.12 R framework
5.3.13 MATLAB
5.3.14 Tableau
5.3.15 Overview of software
5.4 Recent case studies
5.5 Conclusion
References
Section 2: Clinical applications of R-n-R in neuro-oncology
Chapter 6 Survival estimation of brain tumor patients using radiogenomics-based studies
Abstract
Keywords
6.1 Introduction
6.2 Various imaging modalities
6.3 Cancer survival estimation
6.3.1 Overall survival (OS)
6.3.2 Progression-free survival (PFS)
6.3.3 Time to progression (TTP)
6.3.4 Binary lens of SEER studies
6.4 ML techniques
6.5 ML models in brain cancer studies
6.6 Data sources and integration
6.7 Predictive accuracy and validation
6.7.1 Predictive accuracy
6.7.2 Validation
6.7.3 Clinical significance
6.8 Clinical applications and implications
6.9 Limitations and challenges
6.9.1 Heterogeneity of clinical data
6.9.2 Handling missing data
6.9.3 Feature engineering challenges
6.9.4 Addressing bias
6.9.5 Overfitting mitigation
6.9.6 Model interpretability
6.9.7 Generalization to new patients
6.9.8 Anonymization and privacy
6.9.9 Secure data sharing
6.9.10 Ethical considerations
6.9.11 Future directions
6.10 Conclusion
References
Chapter 7 Brain tumor progression analysis: A comprehensive review
Abstract
Keywords
7.1 Introduction
7.2 Literature review
7.3 Brain tumor analysis
7.3.1 Key molecular signaling pathways in brain tumors [9]
7.4 Methods and materials
7.4.1 Pseudoprogression in high-grade glioma
7.4.2 MRI analysis
7.4.3 Adult brain cancer
7.4.4 Origin of brain tumor
7.4.5 Cytokine patterns
7.5 Discussion
7.6 Conclusion
Conflict of interest statement
References
Chapter 8 Integrative data analysis of MGMT methylation and IDH1 mutation in glioblastoma: A comprehensive review
Abstract
Keywords
8.1 Introduction
8.2 Data source and analytical approach
8.2.1 Data source
8.2.2 Analytical approach
8.2.3 Data interpretation
8.3 Analysis of MGMT methylation in GBM
8.4 Analysis of IDH1 mutation in GBM
8.5 Comparative analysis and synthesis of findings
8.5.1 Unveiling the interplay: IDH1 mutation and MGMT methylation in GBM
8.5.2 Converging paths: Similarities and correlations
8.5.3 Diverging paths: Discrepancies and individual effects
8.5.4 Synthesis: A multifaceted picture
8.5.5 Future directions
8.6 Discussion
8.6.1 Improved OS with methylated MGMT
8.6.2 Decoding IDH1 mutations: A beacon for GBM research and treatment
8.7 Conclusion
References
Chapter 9 AI enabled R-n-R for neurooncology: Clinical applications
Abstract
Keywords
9.1 Introduction
9.2 Exploring radiomics and radiogenomics
9.3 Radiogenomic workflow
9.4 Crucial role of genomic studies in advancing cancer research
9.4.1 Feature-based algorithm for radiomics
9.4.2 DL-based algorithm for radiomics
9.5 Brain tumors: Clinical application of AI
9.5.1 Radiomics for glioma grading
9.5.2 Grading of glioma through radiomics
9.6 Radiogenomics
9.6.1 Unlocking glioma mysteries: A radiogenomic odyssey
9.7 Pretreatment assessment
9.7.1 Tumor segmentation
9.7.2 Brain tumor infiltration and extent
9.7.3 Value of prognosis: Survival
9.8 Posttreatment assessment
9.8.1 Pseudoprogression (PSP) in gliomas: A diagnostic challenge
9.8.2 Distinguishing recurrence of tumor from posttreatment changes in gliomas
9.8.3 Radiomics beyond gliomas: Metastasis and primary central nervous system lymphoma (PCNSL) differentiation
9.9 Advancements in pediatric brain tumor characterization through radiomics
9.9.1 Post fossa tumors
9.10 Radiogenomic insights into pediatric brain tumors
9.10.1 Medulloblastoma: Unveiling molecular diversity for targeted treatment
9.11 Recent achievement of radiogenomics
9.12 AI-enabled radiogenomics: Latest developments
9.13 Conclusion
References
Section 3: AI in R-n-R for neuro-oncology: What have we achieved so far?
Chapter 10 AI for application solutions for healthcare services using AI detection and diagnosis of different diseases: A special emphasis on neuro-oncology
Abstract
Keywords
Conflict of interest
10.1 Introduction
10.1.1 An overview of the studies in medical AI
10.2 Healthcare AI use cases in the real world
10.2.1 Automatic arrhythmia analysis using AI
10.2.2 Artificial intelligence-based pediatric illness diagnosis
10.2.3 Using AI-enabled cloud-based IoT, smart skin health monitoring
10.2.4 Mental health and AI
10.2.5 Fetal status evaluation during labor
10.3 The functions of rural medical AI-related techniques
10.3.1 A network of levels of medical AI services
10.3.2 Difficulties and solutions
10.4 Medical precision
10.4.1 Integration between AI and precision medicine in the future
10.5 AI in R-N-R for neuro-oncology
10.5.1 Radiomics
10.5.2 AI in radiogenomics
10.5.3 Clinical challenges and a look ahead: Future scopes
10.6 Advantages and difficulties of AI in hospitals
10.6.1 Save time and resources
10.6.2 Improved diagnostic efficiency
10.6.3 Enhanced patient care
10.6.4 Delivers real-time information
10.6.5 Assists research
10.6.6 Streamlines tasks
10.6.7 Need human supervision
10.6.8 Errors are still conceivable
10.6.9 Overlooking social variables
10.6.10 The threat of data loss
10.6.11 This may lead to unemployment
10.6.12 Lack of personal engagement
10.7 Future trends of AI in healthcare
10.8 Conclusion
References
Chapter 11 Traditional and advanced AI methods used in the area of neuro-oncology
Abstract
Keywords
11.1 Introduction
11.2 Neuro-oncology in the 19th and 20th centuries
11.2.1 Early observations (19th century)
11.2.2 Evolution of neuro-oncology modalities in the 20th century
11.3 Diagnostic imaging techniques in neuro-oncology
11.4 Histopathological analysis in neuro-oncology
11.4.1 Statistical analysis in neuro-oncology clinical studies: Unveiling patterns, predictions, and outcomes
11.4.2 Technological advancements in traditional methods: Elevating efficiency and precision in neuro-oncology
11.5 Advanced methods
11.5.1 Automated brain tumor segmentation using DL
11.5.2 Predictive modeling for glioblastoma survival
11.5.3 NLP for radiology report analysis
11.5.4 AR-assisted brain tumor surgery planning
11.5.5 Integration of multiomics data for glioma subtyping
11.6 Discussion
11.7 Conclusion
References
Chapter 12 AI in radiomics and radiogenomics for neuro-oncology: Achievements and challenges
Abstract
Keywords
12.1 Introduction
12.2 Fundamentals of AI and its components (ML/DL)
12.2.1 Machine learning
12.2.2 Deep learning
12.3 Radiomics and radiogenomics: A brief update in neuro-oncology
12.3.1 Radiogenomics in neuro-oncology
12.4 Achievements: What we have achieved so far?
12.4.1 Improved tumor characterization
12.4.2 Dynamic treatment response assessment
12.4.3 Radiogenomics integration for molecular insights
12.4.4 Enhanced predictive modeling
12.4.5 Survival of brain tumors
12.4.6 Predictions of infiltration and recurrence
12.4.7 Response assessment of brain tumors
12.4.8 Characterization of imaging heterogeneity
12.5 Challenges
12.5.1 Data standardization and availability
12.5.2 Interpretability and explainability
12.5.3 Ethical and legal considerations
12.5.4 Algorithm robustness and generalization
12.5.5 Clinical validation and regulatory approval
12.5.6 Cost and resource implications
12.5.7 Integration into clinical practice
12.6 Conclusion
References
Index

Sanjay Saxena

Dr. Sanjay Saxena is an Assistant Professor in the Department of Computer Science and Engineering at IIIT Bhubaneswar, India. He obtained his Ph.D. from the Indian Institute of Technology (BHU), Varanasi, and completed his postdoctoral research at the Artificial Intelligence in Biomedical Imaging Lab, University of Pennsylvania, USA. Dr. Saxena's primary research area involves developing AI-based methods for brain cancer analysis, with a broader focus on Data Science, Machine Learning, Deep Learning, and the emerging fields of Radiomics and Radiogenomics. His tenure under Prof. Davatzikos at the University of Pennsylvania was instrumental in his significant learning about Radiomics/Radiogenomics. Dr. Saxena is a member of prestigious organisations such as IEEE, the Society of Neuro-Oncology, ACM, and the New York Academy of Science. He has presented his work at globally recognised universities such as like Imperial College London, Stony Brook University, and Vienna University of Technology. and has made substantial scholarly contributions, with several peer-reviewed Journals, conference publications, book chapters, and two books to his name, advancing the intersection of AI and human cancer research.

Affiliations and expertise

Assistant Professor, Department of Computer Science and Engineering, International Institute of Information Technology, Bhubaneswar, India

Jasjit S. Suri

Dr. Jasjit Suri, PhD, MBA, is an innovator, visionary, scientist, and internationally known world leader. Dr Suri received the Director General’s Gold medal in 1980 and Fellow of (i) American Institute of Medical and Biological Engineering, awarded by the National Academy of Sciences, Washington DC, (ii) Institute of Electrical and Electronics Engineers, (iii) American Institute of Ultrasound in Medicine, (iv) Society of Vascular Medicine, (v) Asia Pacific Vascular Society, and (vi) Asia Association of Artificial Intelligence. Dr. Suri was honored with life time achievement awards by Marcus, NJ, USA and Graphics Era University, Dehradun, India. He has published nearly 300 peer-reviewed Artificial Intelligence articles, nearly 2000 Google Scholar Publications, 100 books, and 100 innovations/trademarks leading to an H-index of nearly 100 with about 43,000 citations. He has held positions as chairman of AtheroPoint, CA, USA, IEEE Denver section, Colorado, USA, and advisory board member to healthcare industries and several universities in the United States of America and abroad.

Affiliations and expertise

Chairman, AtheroPoint LLC, USA

Life Sciences

Physical Sciences & Engineering

Social Sciences & Humanities

Health