Biomedical Imaging: Advances in Artificial Intelligence and Machine Learning (Biological and Medical Physics, Biomedical Engineering)

Preface
Contents
Contributors
Abbreviations
1 Artificial Intelligence in Diagnostic Medical Image Processing for Advanced Healthcare Applications
	1.1 Introduction
	1.2 Medical Imaging in Modern Healthcare
		1.2.1 Evolution of Medical Imaging Modalities
		1.2.2 Role of Multiscale Multiparametric Imaging in Healthcare
		1.2.3 Primary Challenges in Medical Imaging: Importance of AI
	1.3 AI in Medical Imaging
		1.3.1 Basics of Artificial Intelligence
		1.3.2 Historical Evolution of AI
		1.3.3 Overall Workflow of AI-Assisted Biomedical Imaging
		1.3.4 Key Techniques and Algorithms in AI for Medical Imaging
	1.4 Applications of AI in Medical Imaging
		1.4.1 Applications of AI in MRI
		1.4.2 Applications of AI in CT
		1.4.3 Applications of AI in PET
		1.4.4 Applications of AI in USG Imaging
		1.4.5 Applications of AI in EIT
		1.4.6 Applications of AI in Optical Microscopy
		1.4.7 Applications of AI in Hybrid Imaging Modalities
	1.5 Benefits and Implications of AI in Medical Imaging
		1.5.1 Improved Accuracy and Efficiency
		1.5.2 Impact on Research and Development in Healthcare
	1.6 Challenges and Considerations
		1.6.1 Ethical and Regulatory Considerations
		1.6.2 Data Privacy and Security
		1.6.3 Integration with Clinical Workflow
	1.7 Conclusions
	References
2 From Pixels to Predictions: Exploring the Role of Artificial Intelligence in Radiology
	2.1 Introduction
	2.2 Medical Image Processing
	2.3 Artificial Intelligence
		2.3.1 Machine Learning
		2.3.2 Deep Learning
		2.3.3 Development of AI-Based Applications in Radiology
		2.3.4 AI Imaging Applications in Radiology
		2.3.5 Challenges and Future of AI in Radiology
		2.3.6 Conclusion
	References
3 Challenges in Accurately Using Artificial Intelligence and Machine Learning in Biomedical Imaging
	3.1 Introduction
	3.2 AI and ML in Biomedical Imaging
		3.2.1 Traditional Machine Learning Algorithms
		3.2.2 Deep Learning Based AI Algorithms
		3.2.3 Convolutional Neural Networks for Medical Imaging Analysis
	3.3 Challenges in Application of AI/ML in Biomedical Imaging
		3.3.1 Intricacy and Data Heterogeneity of Medical Images
		3.3.2 Diverse Tasks in Various Clinical Settings
		3.3.3 Challenges in Curation of Medical Imaging Data
		3.3.4 The Traditional Hypothesis-Driven Research Model Delays Knowledge Dissemination
		3.3.5 Algorithmic Design and Performance Assessment in Medical Imaging
		3.3.6 Challenges in Algorithm Transparency, Validation, and Testing
	3.4 Conclusion
	References
4 Tracing Historical Connections: The Evolutionary Ties of Artificial Intelligence, Confocal Microscopy, and Marvin Minsky (1927–2016)
	4.1 Introduction
	4.2 Marvin Minsky’s Pioneering Work in Artificial Intelligence
	4.3 Minsky’s Contribution to the Development of Confocal Microscopy
	4.4 Conclusion and Future Implications
	References
5 Types of Optical Microscopic Analysis for Cell Death Using Artificial Intelligence
	5.1 Introduction
		5.1.1 Types of AI
	5.2 Types of Microscopy
		5.2.1 Spontaneous Raman Technique
		5.2.2 Fluorescence Techniques
		5.2.3 Coherent Raman Scattering (CARS and SRS) Microscopy
	5.3 Conclusion
	References
6 Applications of Artificial Intelligence in the Analysis of Images of the Oral Cavity for Cancer Detection
	6.1 Introduction
	6.2 AI Applications in Oral Cancer
		6.2.1 Imaging Modalities or Datasets for Screening of Oral Cavity Cancer
		6.2.2 Imaging Data Employed for Detection of Oral Cancer Through Artificial Intelligence Technique
		6.2.3 Traditional Image Processing Approach
		6.2.4 Deep Learning Approach
	6.3 Clinical Applications of AI in Oral Cancer
	6.4 Ethical Concerns of AI
	6.5 Conclusions
	References
7 Leveraging Machine Learning for Advanced Biomedical Imaging: Insights from Speckle Pattern Analysis
	7.1 Introduction
		7.1.1 Speckle Pattern Analysis in Biomedical Imaging
		7.1.2 Advancements in Optical Recognition Through Speckle Pattern Analysis
		7.1.3 Machine Learning in Speckle Analysis
		7.1.4 Applications in Biomedical Imaging
	7.2 Remote Photonic Sensing of Blood Oxygen Saturation
		7.2.1 Motivation and Experimenting
		7.2.2 Data Exploration and Pre-Processing
		7.2.3 Method Overview
		7.2.4 Experimental Results
		7.2.5 Findings and Implications
	7.3 Human Identification via Speckle Patterns
		7.3.1 Motivational Pursuits
		7.3.2 Experimental Setup
		7.3.3 Method Overview
		7.3.4 Experimental Results
		7.3.5 Insights and Recommendations
	7.4 Brain Activity Sensing
		7.4.1 Inspiration and Exploration
		7.4.2 Experiment Setup
		7.4.3 Method Overview
		7.4.4 Research Summary
	7.5 Challenges and Future Perspectives
	7.6 Conclusions
	References
8 Two Photon Fluorescence Integrated Machine Learning for Data Analysis and Interpretation
	8.1 Introduction to Two-Photon Excitation Process
	8.2 TPM Implementation and Data Acquisition
	8.3 TPM Data Preprocessing
	8.4 TPM Data Informative Features Extraction
	8.5 Applications
	8.6 Conclusion
	References
9 Deep Learning in Biomedical Applications of Raman Spectroscopy
	9.1 Introduction
	9.2 Fundamentals of Deep Learning
		9.2.1 Building Blocks of Deep Learning
		9.2.2 How Does the Network Form?
		9.2.3 Is ANN Capable of Solving All Problems?
		9.2.4 The Reason Behind the Real Hype of Convolution
	9.3 Applications of Deep Learning: Spontaneous Raman Spectroscopy
		9.3.1 Disease Diagnosis
		9.3.2 Microbial Cellular Imaging and Classification
		9.3.3 Forensic Applications
		9.3.4 Biochemical Process Analysis
		9.3.5 Environmental Studies
	9.4 Applications of Deep Learning: Surface Enhanced Raman Spectroscopy (SERS)
	9.5 Applications of Deep Learning: Stimulated Raman Spectroscopy
	9.6 Applications of Deep Learning: Coherent Anti-Stokes Raman Spectroscopy (CARS)
	9.7 Conclusions
	References
10 Covid-19 and Pneumonia Detection from Chest X-Ray Images by Deep Learning Model
	10.1 Introduction
	10.2 Existing CNN Models for Covid-19 and Pneumonia Detection from CXR Dataset
		10.2.1 Basic Concept of CNN Model
		10.2.2 Pre-Trained CNN Models
		10.2.3 CNN Frameworks for Alleviating Class Imbalance Problems from CXR Datasets
	10.3 Results and Analysis
		10.3.1 Training Specifications
		10.3.2 Quality Metrics
		10.3.3 Imbalanced CXR Dataset and Its Challenges
		10.3.4 Comparison and Analysis
	10.4 Concluding Remarks
	References
11 Deep Learning Integrated Multiphoton Microscopy
	11.1 Introduction
	11.2 Application of  Two-Photon Fluorescence Microscopy in Image Analysis
	11.3 Application of Multiphoton Microscopy (MPM) Using Deep Learning
	11.4 Conclusion
	References
12 U-Net: A Versatile Deep Learning Architecture for Multi-Disease Detection
	12.1 Introduction
	12.2 Overview of U-Net and Its Applicability in Medical Imaging
		12.2.1 Loss Functions and Optimization
		12.2.2 Evaluation Metrics
		12.2.3 U-Net Variants and Modifications
		12.2.4 Data Augmentation Strategies to Improve U-Net Performance
	12.3 Summary of Important Datasets
	12.4 U-Net Applications in Multi-Disease Detection
		12.4.1 U-Net for Tumor Segmentation in Brain MRI Scans
		12.4.2 U-Net for Lung Nodules Detection Using CT Scans
		12.4.3 U-Net for Liver Lesion Segmentation in Abdominal MRI Scans
		12.4.4 Transfer Learning in U-Net
	12.5 Conclusion and Future Directions
	References
13 Memristor-Based Neuromorphic Computing and Artificial Neural Networks for Computer Vison and AI—Applications
	13.1 Introduction
		13.1.1 Memristors
		13.1.2 Neuromorphic Computing
	13.2 Applications of Neuromorphic Computing
		13.2.1 Medical Image Analysis and Computer Vision
		13.2.2 Neuromorphic Prosthetics and Brain-Machine Interfaces
		13.2.3 Disease Diagnosis and Prediction
		13.2.4 Drug Discovery and Development
		13.2.5 Brain Disorders and Cognitive Health
		13.2.6 Healthcare Robotics and Assistance
		13.2.7 Object Recognition
		13.2.8 Augmented Reality (AR) and Virtual Reality (VR)
	13.3 Conclusion
	References
14 Generations of CT—the Evolution and Future Prospects
	14.1 Introduction
		14.1.1 First-Generation CT
		14.1.2 Second Generation CT
		14.1.3 Third Generation CT
		14.1.4 Fourth-Generation CT
		14.1.5 Fifth Generation CT
		14.1.6 Sixth Generation CT
	14.2 Multi-Slice/Multi-Detector CT
	14.3 Technological Advancements
	14.4 How Does the Future Look like?
	14.5 Conclusion
	References
Index