Proceedings of 2023 International Conference on Medical Imaging and Computer-Aided Diagnosis (MICAD 2023): Medical Imaging and Computer-Aided Diagnosis (Lecture Notes in Electrical Engineering, 1166)

✍ Scribed by Ruidan Su (editor), Yu-Dong Zhang (editor), Alejandro F. Frangi (editor)

• Publisher: Springer
• Year: 2024
• Tongue: English
• Leaves: 414
• Category: Library

✦ Synopsis

This book covers virtually all aspects of image formation in medical imaging, including systems based on ionizing radiation (x-rays, gamma rays) and non-ionizing techniques (ultrasound, optical, thermal, magnetic resonance, and magnetic particle imaging) alike. In addition, it discusses the development and application of computer-aided detection and diagnosis (CAD) systems in medical imaging. This book includes the state-of-the-art research of computer-aided diagnosis systems with aritificial intelligence.

Given its coverage, the book provides both a forum and valuable resource for researchers involved in image formation, experimental methods, image performance, segmentation, pattern recognition, feature extraction, classifier design, machine learning / deep learning, radiomics, CAD workstation design, human–computer interaction, databases, and performance evaluation.

✦ Table of Contents

Preface
Organization
Contents
Medical Imaging
Detecting Pulmonary Lesions in Low-Prevalence Real-World Settings Using Deep Learning
1 Introduction
2 Problem
3 Proposed Software
3.1 Model Architecture
3.2 Training Data
4 Methodology
4.1 Objectives
4.2 Data Source and Demographic Data
4.3 Ground Truth
4.4 Assessment
4.5 Statistical Analysis
5 Results
6 Discussion
6.1 Limitations
7 Conclusions
References
Evaluation of Left Main Bifurcation Ostial Area by Main Vessel Intravascular Ultrasound
1 Introduction
2 Methods
2.1 Study Population
2.2 IVUS Image Acquisition and Analysis
2.3 Analysis from MB Pullback and SB Pullback
2.4 Subgroup Analysis of Pullbacks
2.5 Statistical Analysis
3 Results
3.1 Baseline Clinical Characteristics
3.2 IVUS Image Analysis Results
3.3 Motion Artifacts and Placement of Stent
4 Discussion
4.1 Clinical Implications
4.2 Limitations of the Study
5 Conclusion
References
End-to-End Autoencoding Architecture for the Simultaneous Generation of Medical Images and Corresponding Segmentation Masks
1 Introduction
2 Proposed Method
2.1 The Hamiltonian VAE
2.2 Modeling the Joint Distribution for the Simultaneous Medical Images and Masks Generation
3 Experiments
3.1 Datasets
3.2 Training Settings
3.3 Evaluation Results
4 Conclusion
References
Evaluation of Randomized Input Sampling for Explanation (RISE) for 3D XAI - Proof of Concept for Black-Box Brain-Hemorrhage Classification
1 Introduction
1.1 Motivation
1.2 Prior Work
1.3 Contributions
2 Methods and Materials
2.1 Overview
2.2 Dataset
2.3 Randomized Input Sampling for Explanation (RISE)
2.4 3D Rise
2.5 Explainability Quality Assessment
2.6 Explainability Quantitative Assessment
3 Results
3.1 Empirical Tuning of Parameters
3.2 Mask Combination Experiments
3.3 Relationship Between Pathology Distribution and Heatmap Quality
3.4 Test Set Results
4 Discussion and Outlook
References
Multi-task Learning Approach for Unified Biometric Estimation from Fetal Ultrasound Anomaly Scans
1 Introduction
2 Methods
2.1 Multi-task Learning Network
2.2 Estimating the Bio-Parameters
3 Data and Experiments
3.1 Dataset
3.2 Experiments
4 Results and Discussion
5 Conclusion
References
Energy-Efficient 3D Convolution Using Interposed Memory Accelerator eXtension 2 for Medical Image Processing
1 Introduction and Background
2 IMAX2 and U-Net
2.1 IMAX2
2.2 U-Net and Its Bottlenecks
3 Implementation and Results
3.1 Implementing 3D CNN on IMAX2
3.2 Results
4 Conclusion and Future Prospects
References
Grey Level Texture Features for Segmentation of Chromogenic Dye RNAscope from Breast Cancer Tissue
1 Introduction
1.1 RNAscope Staining
1.2 RNAscope Segmentation
2 Related Work
3 Method
3.1 Data Acquisition
3.2 Expert Annotation of RNAscope Transcripts
3.3 Candidate Selection
3.4 Feature Extraction
4 Results
4.1 Feature Analysis
5 Conclusion
References
Predicting Central Cervical Lymph Node Metastasis of Papillary Thyroid Carcinomas Using Multi-view Ultrasound Images
1 Introduction
2 Method
2.1 ROI Extraction
2.2 Multi-view Classification Model
2.3 Focal Loss
3 Experiments
3.1 Dataset and Experiment Setting
3.2 Evaluation Metrics
3.3 Result Analysis
4 Conclusion
References
Modified Technique “Modus Spirdonov 1” in Fine-Needle Aspiration Biopsy Under Ultrasonographical Control
1 Introduction
2 Materials and Methods
3 Results
4 Discussion
5 Conclusion
References
Customized Position with a Breast Pad for MDCT – A Single-Institution Experience for Breast Cancer Staging
1 Introduction
2 Materials and Methods
3 Results
4 Conclusion
References
Self-supervised Probe Pose Regression via Optimized Ultrasound Representations for US-CT Fusion
1 Introduction
1.1 Problem Definition
1.2 Generative Models for Ultrasound Imaging
2 Method
2.1 Differentiable Ultrasound Rendering Module
2.2 Bridging the Gap Using Generative Models
2.3 Pose Regression Network
2.4 Overall Pipeline
3 Experiment
4 Results and Discussion
5 Conclusion
References
Machine Learning and AI Approaches for Classifying Primary Brain Tumours Using Conventional MRI Scans
1 Introduction
1.1 Problem Definition
1.2 Methodology and Data Sets
2 Classification of Tumour Types
2.1 Data Preprocessing
2.2 Classification – Attempt 1
2.3 Classification – Attempt 2
3 Discussion
4 Future Directions
5 Conclusion
References
Ret2Ret: Retinal Blood Vessel Extraction via Improved Pix2Pix Image Translation
1 Introduction
2 Methodology
2.1 Pix2Pix GAN for Image to Image Translation
2.2 Ret2Ret: Modified U-Net in Generator for Vessel Extraction
2.3 PatchGAN Based Discriminator
2.4 Training Protocol
3 Experiments and Results
3.1 Benchmark Databases
3.2 Image Pre-processing
3.3 Experimental Results
4 Conclusion
References
A Data Augmentation Approach to Enhance Breast Cancer Segmentation
1 Introduction
2 Methodology
3 Experimental Design
3.1 Data
3.2 Configuration
3.3 Evaluation Metrics
4 Results and Observations
5 Conclusion and Future Work
References
PB-FELTuCS: Patch-Based Filtering for Enhanced Liver Tumor Classification and Segmentation
1 Introduction
2 Related Works
3 Methodology
3.1 Network Architecture
3.2 Patch-Based Filtering Algorithm
3.3 Training Algorithm
4 Experiments
4.1 Liver Segmentation
4.2 Tumor Classification
4.3 Tumor Segmentation
4.4 Ablation Study
5 Conclusion
References
AlexNet for Image-Based COVID-19 Diagnosis
1 Background of AlexNet
2 Introduction of COVID-19
3 Image-Based COVID-19 Diagnosis
4 AlexNets Improve Diagnosis Accuracy
5 Challenges of AlexNet
6 Conclusion
References
Computer-Aided Detection/Diagnosis
Diagnostic Genes Identification and Molecular Classification Patterns Based on Oxidative Stress-Related Genes in Ischemic Stroke
1 Introduction
2 Materials and Methods
2.1 Data Acquisition
2.2 Gene Set Variation Analysis (GSVA) and Differentially Expressed Genes Analysis
2.3 Selection of Characteristic Genes
2.4 Evaluation of Immune Cell Infiltration
2.5 Consensus Clustering
2.6 Weighted Gene Co-Expression Network Analysis (WGCNA) and Enrichment Analysis
2.7 Seurat Object Creation and Data Preprocessing
2.8 Pseudotime Analysis and Cell Communication Analysis
2.9 Statistical Analysis
3 Results
3.1 GSVA Analysis and Identification of DEGs
3.2 Selection of Signature Genes and Establishment of Diagnostic Model
3.3 Immune Cell Infiltration Analysis
3.4 Identification of Oxidative Stress-Related Genes Subtypes in Ischemic Stroke
3.5 Construction of Co-expression Networks and Analysis of Enrichment
3.6 Identification of Cell Clusters Based on scRNA-Seq Data
3.7 Pseudotime Analysis of Endothelial Cells
3.8 Cell–Cell Communication
4 Discussion
5 Conclusion
References
Identification of a Novel Model for Predicting the Prognosis and Immune Response Based on Genes Related to Ferroptosis and Disulfidptosis in Liver Hepatocellular Carcinoma
1 Introduction
2 Materials and Methods
2.1 TCGA Transcriptome and Clinical Data Download and Pre-organization
2.2 Construction of a Prognostic Risk Model
2.3 Construction of a Nomogram and Functional Enrichment Analysis
2.4 Correlation Analysis with Immune Infiltration
3 Results
3.1 Expression of Disulfidptosis and Ferroptosis Relevant Genes in LIHC
3.2 Construction of the Prognostic Model
3.3 Construction of Nomogram and Analysis of Functional Enrichment
3.4 Immune Infiltration Analysis
4 Discussion
5 Conclusions
References
Progress of Intelligent Diagnosis via Multiple Brain Features in Alzheimer’s Disease
1 Introduction
2 Overview of ID for AD
2.1 Clinical Scales
2.2 Gene and CSF Biomarkers
2.3 Brain Neuroimaging
2.4 Text Mining
2.5 Combined Features
3 Conclusions
References
Classification of Children with/without Autism Spectrum Disorder Using Speech Signal
1 Introduction
2 Dataset
3 Experiments
3.1 Usage of PCA
3.2 Direct Truncation of Feature Set
4 Conclusions
References
Identification of FECG from AECG Recordings using ICA over EMD
1 Introduction
2 Proposed Method for fECG Extraction from aECG
3 Results and Discussion
3.1 Results
3.2 Performance Evaluation and Discussion
4 Conclusions
References
Classification of Pathological Speech in Speakers with Cleft Palate: Decision Tree Approach
1 Introduction
2 Materials and Method
2.1 Design/Patients
2.2 Recording
2.3 Corpus
2.4 Extraction of Acoustic Parameters
3 Utilizing Decision Trees in Our Classification
3.1 Learning Phase
3.2 Classification Phase
4 Discussion
5 Conclusion
References
Towards Seamless Surgical Guidance: A Robust Marker-Based Multi-camera AR Navigation System with Advanced Calibration and Detection Techniques
1 Introduction
2 Multicamera AR Navigation System
2.1 System Overview
2.2 Marker Pose Estimation
2.3 Multicamera Collaborative Registration System
3 Experiments
3.1 Evaluation of the Pose Estimation System
3.2 Evaluation of Registration System on Simulation Platform
3.3 Evaluation for CT-Guided Interventional Surgery
4 Conclusion
References
Machine Learning and Deep Learning
Deep Learning Approaches for Automated Classification of Muscular Dystrophies from MRI
1 Introduction
2 Materials and Methods
2.1 Patient MRI Data
2.2 Deep Learning
2.3 Machine Learning
3 Results
4 Discussion
References
Identification of Hub Biomarkers and Immune Cell Infiltration Characteristics in Ulcerative Colitis by Bioinformatics Analysis and Machine Learning
1 Introduction
2 Materials and Methods
2.1 Download and Pre-Processing of the Ulcerative Colitis Dataset
2.2 Construction of Weighted Gene Co-expression Network
2.3 DO Analysis, GO Analysis, KEGG Enrichment Analysis
2.4 Machine Learning Methods Screen for Genes Characteristic of Ulcerative Colitis
2.5 Identifying and Verifying Feature Genes
2.6 GSEA Analysis of the Feature Genes and Immune Cell Infiltration
3 Results
3.1 Screening of DEGs
3.2 Construction of Weighted Gene Co-Expression Network and Identification of Key Modules
3.3 Acquisition and Enrichment Analysis of Intersection Genes
3.4 Selection of Disease Feature Genes
3.5 Validation of Feature Gene Expression Levels and Construction of Receiver Operating Characteristic Curves
3.6 Gene Set Enrichment Analysis
3.7 Immune Infiltration Analysis
4 Discussion
5 Conclusions
References
TMN: An Efficient Robust Aggregator for Federated Learning
1 Introduction
1.1 Related Works
2 Methodology
3 Data and Experiments
3.1 Experiments
3.2 Dataset
4 Results and Discussion
4.1 Performance
4.2 Scalability
5 Conclusion and Future Work
References
Research Progress of Deep Learning in Thyroid Nodule Imaging Examination
1 Introduction
2 Application of Deep Learning in Medical Imaging
2.1 Convolutional Neural Networks
2.2 Transformer
2.3 Workflow
3 Research Based on Different Imaging
3.1 Public Datasets
3.2 Application Based on Ultrasound
3.3 Application Based on CT
3.4 Application Based on MRI
3.5 Combined with the Application of Multiple Imaging
4 Conclusion and Prospect
References
Comparing Different Deep-Learning Models for Classifying Masses in Ultrasound Images
1 Introduction
2 Background and Related Work
3 Methodology
3.1 Deep Learning-Based Segmentation
3.2 Transfer Learning-Based Classification
4 Implementation
4.1 Dataset
4.2 Implementation of the DL-Based Segmentation
4.3 Implementation of the TL-Based Classification
5 Results and Discussion
6 Conclusions and Recommendations
References
RNN-Based Multiple Instance Learning for the Classification of Histopathology Whole Slide Images
1 Introduction
2 Related Work
2.1 MIL Method Applied in WSI Classification
2.2 RWKV and RNN-Like Architectures
3 Method
3.1 WSI Patching and Tokenization
3.2 RWKV-Based Multiple Instances Learning
4 Experiments and Results
4.1 Dataset and Experiment Setup
4.2 Results of WSI Classification
4.3 Ablation Study
5 Conclusion
References
Machine Learning for Image Denoising: A Review
1 Introduction
2 Background of Image Denoising
3 Image Noise Types
3.1 Gaussian Noise
3.2 Salt-And-Pepper Noise
3.3 Poisson Noise and Other Types
4 Machine Learning Theories and Techniques
5 Deep Learning Architectures
6 Evaluation Metrics
7 Conclusion
References
Deep Learning for Image Classification: A Review
1 Introduction
2 Background of Image Classification
3 Traditional Approaches to Image Classification
4 Deep Learning Approaches to Image Classification
5 Common Image Classification Datasets
6 Conclusion
References
Others
Survival Modeling of Disease Consequences and Post-disease Syndromes
1 Introduction
1.1 Polio Outbreak
2 Survival Model
3 Czech Polio Model Assumptions
3.1 Czech Polio Model
4 Results
5 Conclusions
References
Navigating Privacy and Security Challenges in Electronic Medical Record (EMR) Systems: Strategies for Safeguarding Patient Data in Developing Countries – A Case Study of the Pacific
1 Introduction
1.1 Related Work
1.2 Problem Statement
2 Privacy and Security Challenges in EMR Systems (PATIS)
3 Case Study: The Pacific Developing Countries
4 Methodology
5 Proposal for Optimal Methodology and Functioning
5.1 Implementing Affordable Cloud-Based Solutions
5.2 Proposed Cloud-Based Implementation Prototyping
6 Discussion
7 Conclusion
References
Construction of Precision Medical Model Based on Electronic Medical Records
1 Introduction
2 The Construction of Precision Medical Model Based on Electronic Medical Records
2.1 Data Layer
2.2 Data Analysis and Governance Layer
2.3 Application Service Layer
3 Discussion on Model-Based Precision Medical Service Optimization
3.1 Legal Construction
3.2 Electronic Medical Record Data Utilization Optimization
3.3 Improvement of Precision Medical Service Model
References
Research and Application of an Intelligent Anesthesia Drug Management System Based on Internet of Things Technology
1 Introduction
2 Construction of Intelligent Medication Management System
2.1 System Architecture
2.2 System Function
3 Operational Process of Intelligent Medication Drug Management System
3.1 Inventory Management
3.2 Statistical Reports
3.3 Requisition Management
3.4 Surgery Scheduling
3.5 Medication Plans
3.6 Medicine Cabinet Configuration
3.7 Configuration Center
4 Application Effect of the Intelligent Medication Management System
4.1 Shortening Drug Dispensing Time
4.2 Achieving Electronic Ledger Management
4.3 Enabling Electronic Shift Turnovers
5 Conclusion
References
Author Index