Proceedings of 2021 International Conference on Medical Imaging and Computer-Aided Diagnosis (MICAD 2021): Medical Imaging and Computer-Aided ... Notes in Electrical Engineering, 784)

Preface
Organization
	General Chair
	General Co-chair
	Program Chairs
	Publication Chair
	Supporting Academic Organizations
	Technical Program Committee
Contents
Medical Imaging
A Dual Supervision Guided Attentional Network for Multimodal MR Brain Tumor Segmentation
	1 Introduction
	2 Method
		2.1 Encoder and Decoder
		2.2 Dual Attention Fusion Block
		2.3 Dual Supervision Strategy
		2.4 The Choice of Loss Function
	3 Experiments
		3.1 Dataset and Implementation Details
		3.2 Evaluation Metrics
		3.3 Experiment Results
	4 Conclusion
	References
Three-Dimensional Image Reconstruction of Murine Heart Using Image Processing
	1 Introduction
	2 Related Work
	3 Method
	4 Experimental Results
	5 Conclusions
	References
Identifying Melanoma in Lesion Images Using Cycle-Consistent Adversarial Networks-Based Data Augmentation
	1 Introduction
	2 Related Work
		2.1 Previous Study on Melanoma Image Classification
		2.2 Data Augmentation
	3 Methodology
		3.1 GAN
		3.2 CycleGAN Image Transformation
		3.3 Melanoma Detection Framework Based on Data Augmentation
	4 Experimental Results
		4.1 Dataset
		4.2 Quantitative Comparisons Between Different Data Augmentation Methods
		4.3 Quantitative Comparison Using Different Network Architectures
	5 Conclusion
	References
Ensembling Learning for Automated Detection of Diabetic Retinopathy
	1 Introduction
	2 Methodology
		2.1 Ensemble Learning Framework
		2.2 Deep Convolutional Networks Models
		2.3 Handcrafted Features
	3 Experimental Results
		3.1 Dataset
		3.2 Implementation Details
		3.3 Evaluation Metrics
		3.4 Quantitative Evaluation
	4 Conclusion
	References
A Fully Automated End-to-End Process for Fluorescence Microscopy Images of Yeast Cells: From Segmentation to Detection and Classification
	1 Introduction
	2 Data
	3 End-to-End Process
	4 Experimental Design
	5 Results and Discussion
	6 Conclusion and Future Work
	References
Glioblastoma Multiforme Patient Survival Prediction
	1 Introduction
	2 BraTS Dataset
	3 Survival Prediction Methodology
		3.1 Predictors and Parameter Tuning
		3.2 Prognosis Using Features
	4 Results and Discussions
		4.1 Image-Based Feature Prediction
		4.2 Radiomics Feature-Based Prediction
		4.3 Discussions
	5 Conclusion
	References
Virtual Reality Application for Laparoscope in Clinical Surgery Based on Siamese Network and Census Transformation
	1 Introduction
	2 Proposed Approach
		2.1 Approach Overview
		2.2 Data Collection and Pre-processing
		2.3 Traditional Sparse Tracking and Reconstruction
		2.4 Densification and Cluster BA
		2.5 Keyframe’s Depth Map Reconstruction
		2.6 Depth Maps Alignment of Keyframes
		2.7 VR System for Clinical Surgery and Medical Training
	3 Result and Discussion
		3.1 Benchmark Hardware and Compared Methods
		3.2 Quantitative Evaluation
	4 Results and Discussion
	References
Analyzing CT Scan Images Using Deep Transfer Learning for Patients with Covid-19 Disease
	1 Introduction
	2 Methodology
	3 Tensor Flow and ImageNet
	4 Convolutional Neural Networks
	5 Deep Transfer Learning (DTL)
	6 Result and Discussion
		6.1 Performance Evaluation
		6.2 Classification of Diseases
	7 Conclusion
	8 Compliance with Ethical Standards
	References
Geometrically Matched Multi-source Microscopic Image Synthesis Using Bidirectional Adversarial Networks
	1 Introduction
	2 Methodology
		2.1 Preliminary Background
		2.2 Model Architecture
		2.3 Loss Functions
		2.4 Geometric Matching Index
	3 Experiments
		3.1 Dataset and Preprocessing
		3.2 Model Parameters and Training
		3.3 Experimental Results
	4 Conclusion
	References
Color-Based Fusion of MRI Modalities for Brain Tumor Segmentation
	1 MRI Segmentation
	2 Fusion of MRI Modalities
	3 Experimental Evaluation
		3.1 Comparison with 2D/3D U-Nets
		3.2 Comparison with a Single Modality
		3.3 Ablation Study
	4 Conclusion
	References
Quantification of Epicardial Adipose Tissue in Low-Dose Computed Tomography Images
	1 Introduction
	2 Data
	3 Experimental Setup
		3.1 Pericardium Interior Region Segmentation
		3.2 Postprocessing Calibration for LDCT
		3.3 EAT Quantification in LDCT
	4 Results
		4.1 Pericardium Interior Region Segmentation
		4.2 Postprocessing Calibration for LDCT
		4.3 EAT Quantification in LDCT
	5 Discussion
	References
Modulated Rotating Orthogonal Polarization Parametric Imaging, A Preliminary Study
	1 Introduction
	2 Experiment Setup and Data Processing
	3 Results and Discussion
	4 Conclusion
	References
Evaluating Mobile Tele-radiology Performance for the Task of Analyzing Lung Lesions on CT Images
	1 Introduction
	2 Materials and Methods
		2.1 Patient Image Data
		2.2 Experimental Design
		2.3 Statistical Analysis
	3 Results
	4 Discussion
	5 Conclusions
	References
Learning Transferable Features for Diagnosis of Breast Cancer from Histopathological Images
	1 Introduction
	2 Related Works
	3 Methodology
		3.1 Stain Normalisation Techniques
		3.2 Data Augmentation
		3.3 Choice of Off The-Shelf Feature Extractors
		3.4 Proposed Framework Architecture
	4 The Experiments and Their Results
	5 Conclusion
	References
Improving Topology Consistency of Retinal Vessel Segmentation via a Double U-Net with Asymmetric Convolution
	1 Introduction
	2 Double U-Net for Retinal Vessel Segmentation
		2.1 Dense Atrous U-Net with Salient Computing
		2.2 Asymmetric Convolution Module
		2.3 Weighted Binary Cross Entropy Loss
	3 Experiment
		3.1 Dataset
		3.2 Evaluation Metrics
		3.3 Ablation Study
		3.4 Comparison Results
	4 Conclusion
	References
The CT Liver Image Segmentation Based on RTV and GMM
	1 Introduction
	2 Threshold Segmentation Method
	3 Relative Total Variation for Image Denoising
	4 Gaussian Mixture Model Segmentation
	5 Segmentation Results and Conclusions
	References
Automated Gland Detection in Colorectal Histopathological Images
	1 Introduction
	2 Related Works
	3 Data and Method
		3.1 The Warwick-QU Dataset Image
		3.2 Pre-processing Image Dataset
		3.3 Evaluation Metrics
		3.4 Proposed Model
	4 The Experiment and Its Results
	5 Conclusion
	References
Ultrasonic Image Segmentation Algorithm of Thyroid Nodules Based on DPCNN
	1 Introduction
	2 DPCNN Model and Its Characteristics
	3 Coarse Segmentation of Thyroid Nodules Based on DPCNN
	4 Coarse Localization of Nodules Based on Regional Expansion Method
	5 Accurate Segmentation of Nodule Coarse Positioning Image
	6 Experimental Results and Analysis
	7 Concluding Remarks
	References
Computer-Aided Detection/Diagnosis
Information Technologies in Complex Reconstructive Maxillofacial Surgery
	1 Introduction
	2 Computer Simulation and Additive Technologies in Maxillofacial Surgery
	3 Conclusion
	References
Machine Learning-Based Imaging in Connected Vehicles Environment
	1 Introduction
	2 Current Medical Imaging Industry
	3 Environment of Clinical Medical Image Acquisition
	4 Introduction to Connected Vehicles
	5 Medical Imaging in Connected Vehicles
	6 Challenges of Imaging in Connected Vehicles Environment
	7 Machine Learning for CVE
	8 Conclusion
	References
Preliminary Considerations on the Design of Multi-layered Bone Scaffold for Laser-Based Printing
	1 Introduction and Literature Review
	2 Materials and Methods
		2.1 Multi-layered Porous Scaffold
		2.2 Design and Manufacture Procedure for Multi-layered Scaffold
	3 Results and Discussion
	4 Conclusions
	References
Two-Stage Convolutional Neural Network for Knee Osteoarthritis Diagnosis in X-Rays
	1 Introduction
		1.1 Our Approaches
		1.2 Contributions
		1.3 More Related Works
	2 Methods
		2.1 Data Preprocessing
		2.2 KneeDetnet for Knee Joint Localization
		2.3 User-Friendly Assessment: KLnet for Knee OA
	3 Results
		3.1 Datasets
		3.2 Experimental Results and Analysis of the Knee Joint Localization
		3.3 Experimental Results and Analysis of the Knee OA Diagnosis
	4 Conclusions
	References
The Art-of-Hyper-Parameter Optimization with Desirable Feature Selection
	1 Introduction
		1.1 Hyper-parameter Optimization (HPO)
		1.2 Model Selection
		1.3 The Common Optimization Strategy
	2 Result and Discussion
		2.1 Finest Features in Descending Approach - Top Twenty
		2.2 Method Obtained in Tuning ML Algorithms
		2.3 Discussion
	3 Conclusion
	References
Data Augmentation for Breast Cancer Mass Segmentation
	1 Introduction
	2 New Data Augmentation Based on Multiple Acquisition Modeling
		2.1 Realistic Transformation Model Based on Image Meshing and Registration
		2.2 Reduced Model of Realistic Transformations
		2.3 Realistic Data Augmentation Model
	3 Contribution of Data Augmentation in Deep Learning Based Segmentation
	4 Conclusions
	References
Dual-Attention Network for Acute Pancreatitis Lesion Detection with CT Images
	1 Introduction
	2 Related Work
		2.1 Detection Architecture
		2.2 Attention Mechanism
	3 Proposed Method
		3.1 Backbone
		3.2 Channel-Wise Attention
		3.3 Spatial Attention
	4 Experiment
		4.1 Dataset
		4.2 Experiment Result
		4.3 Ablation Study
	5 Conclusion
	References
Measurement of Q Factor from Two Dimensional Images of Osteoarthritic Knee Braces
	1 Introduction
		1.1 Types of Braces [3, 4]
		1.2 Finite Element Method
	2 Proposed Solution
		2.1 System Architecture
		2.2 Equations Involved in Analysis
		2.3 Generating Models from 2D Mesh
	3 Discussion
	References
Machine Learning and Deep Learning
2Be3-Net: Combining 2D and 3D Convolutional Neural Networks for 3D PET Scans Predictions
	1 Introduction
	2 Method
	3 Experiments
		3.1 Experimental Setup
		3.2 Experiment 1: Ability to Exploit Spatial Information
		3.3 Experiment 2: Prediction of Clinical Outcomes
	4 Discussion
	5 Conclusion
	References
Covid-19 Chest CT Scan Image Classification Using LCKSVD and Frozen Sparse Coding
	1 Introduction
	2 Data Set
	3 Sparse Representation
	4 Methodology
	5 Results and Conclusion
	References
A Hybrid Deep Model for Brain Tumor Classification
	1 Introduction
	2 Methodology
		2.1 Deep Networks
		2.2 Proposed Classification Framework
		2.3 Train and Validation Model
	3 Result and Discussion
		3.1 Dataset Description and Its Pre-processing
		3.2 Learning Parameters and Model Structure
		3.3 Experimental Matrices
		3.4 Comparison With State-of-the-Art Deep Models
	4 Conclusion
	References
A Systematic Literature Review of Machine Learning Applications for Community-Acquired Pneumonia
	1 Introduction
	2 Methodology
		2.1 Research Questions
		2.2 Searching and Screening
		2.3 Classification and Data Extraction
	3 Results
		3.1 Diagnosis
		3.2 Outcome Prediction
		3.3 ICU Admission Prediction
		3.4 CAP Treatment
	4 Discussion and Conclusion
	References
Photograph to X-ray Image Translation for Anatomical Mouse Mapping in Preclinical Nuclear Molecular Imaging
	1 Introduction
	2 Methodology
		2.1 Data Collection and Preprocessing
		2.2 Modelling and Performance Evaluation
	3 Results
	4 Conclusion and Future Work
	References
Active Strain-Statistical Models for Reconstructing Multidimensional Images of Lung Tissue Lesions
	1 Introduction
	2 Computer Model of Contour Selection
	3 Analysis of the Active Form Model
	4 Conclusion
	References
A New Content-Based Image Retrieval System for SARS-CoV-2 Computer-Aided Diagnosis
	1 Introduction
	2 Related Work
	3 Proposal
		3.1 Motivation
		3.2 System Architecture
	4 Experiments
		4.1 Datasets
		4.2 Experimental Design
		4.3 Results
	5 Conclusion
	References
Dysplasia Grading of Colorectal Polyps Through Convolutional Neural Network Analysis of Whole Slide Images
	1 Introduction
	2 Related Work
	3 Dataset
	4 Method
	5 Results
		5.1 Patches Normalization
		5.2 Study on Patches Resolution for WSI Classification
		5.3 WSI Classification with 600m Patches
	6 Conclusion
	References
Deep YOLO-Based Detection of Breast Cancer Mitotic-Cells in Histopathological Images
	1 Introduction
	2 Methodology for the Proposed Work
		2.1 Target DataSet
		2.2 Data Pre-processing
		2.3 Data Preparation
		2.4 Anchor Boxes Choice
		2.5 Accuracy Metric of Detector
		2.6 Proposed Architecture
	3 The Experiment and Its Results
	4 Conclusions
	References
Others
Promoting Cardiovascular Health Using a Recommendation System
	1 Introduction
	2 The Evolution of Case-Based Reasoning
	3 Methodology
		3.1 The Proposed System
		3.2 Case-Based Reasoning Steps
	4 The Developed System
	5 Conclusions and Future Work
	References
Unsharp Masking with Local Adaptive Contrast Enhancement of Medical Images
	1 Introduction
	2 Algorithms Description
	3 Experimental Results
	4 Conclusion
	References
Building a COVID-19 Literature Knowledge Graph Based on PubMed
	1 Introduction
	2 Building Methods
		2.1 Named Entity Recognition
		2.2 Validation of BERT-BiLSTM-CRF
		2.3 Author Name Disambiguation
	3 CLKG Construction Process
	4 CLKG Visualization
	5 Conclusion
	References
Moving Target Tracking Algorithm Based on Color Space Distribution Information
	1 Moving Target Extraction
	2 Target Tracking
		2.1 Establishment of Target Color Space Distribution Model
	3 Target Tracking
		3.1 Update of Color Area Objects
	4 Conclusion
	References
Predicting Neurostimulation Responsiveness with Dynamic Brain Network Measures
	1 Introduction
	2 Methodology
		2.1 DataSet
		2.2 Region of Interest
		2.3 Multilayer Community Detection
		2.4 Dynamic Network Statistics
		2.5 Dynamic and Static Predictive Parameters
	3 Results
		3.1 Statistical Differences of DFC Characteristics
		3.2 Prediction of Neurostimulation Responsiveness
	4 Discussion and Conclusions
	References
Visualization of Continuous and Pulsed Ultrasonic Propagation in Water
	1 Introduction
	2 System and Theory
		2.1 Polarization Imaging Optical Path
		2.2 Stroboscopic System
		2.3 Theory of Sound Field Visualization
		2.4 Simulation of Sound Field Propagation
	3 Results and Discussion
		3.1 Sound Field Simulation
		3.2 Visualization of Continuous Ultrasound
		3.3 Visualization of Pulsed Ultrasound
	4 Conclusion
	References
An Infrared Imaging Method that Uses Modulated Polarization Parameters to Improve Image Contrast
	1 Introduction
	2 Acquisition of Polarization Parameter Images
		2.1 Infrared Polarization Parameter Imaging System
		2.2 Sample Preparation
		2.3 Measurement Procedure of the Experiment
		2.4 Theoretical Model of the Imaging System
	3 Experimental Results and Analysis
		3.1 Comparison of the Two Processing Models
		3.2 Polarization Parametric Imaging of Subsurface Structure
	4 Conclusion
	References
The Overview of Medical Image Processing Based on Deep Learning
	1 Introduction
	2 Analysis of Medical Image Research
	3 Application of Deep Learning in Medical Image Analysis
		3.1 Medical Image Classification
		3.2 Medical Image Detection
		3.3 Medical Image Segmentation
		3.4 Medical Image Registration
	4 Conclusion
	References
Typical Fault Classification and Recognition of Photovoltaic Modules Based on Deep Learning and Thermal Imaging Picture Processing
	1 Introduction
	2 Typical Fault Classification and Analysis of Photovoltaic Modules
		2.1 Heat Spot
		2.2 Whole Component Failure
		2.3 Strip Battery Malfunction
		2.4 Junction Box Damaged
	3 Inspection System Framework
		3.1 The Structures
		3.2 YOLOv5 Principle
	4 Environment and Models
		4.1 Environment Setting
		4.2 Model Training
	5 Experimental Analysis
		5.1 Experimental Data
		5.2 Experimental Environment
		5.3 Evaluation Criterion
	6 Conclusion
	References
An Obstacle Avoidance Method for Agricultural Plant Protection UAV Based on the Fusion of Ultrasonic and Monocular Vision
	1 Introduction
	2 Common UAV Obstacle Avoidance Methods
		2.1 Ultrasonic Obstacle Avoidance Method
		2.2 Obstacle Avoidance Method of Lidar
		2.3 Millimeter Wave Radar Obstacle Avoidance Method
		2.4 Obstacle Avoidance Method of Machine Vision
	3 UAV Obstacle Avoidance Algorithm Flow Based on the Fusion of Ultrasonic and Monocular Vision
	4 UAV Obstacle Avoidance Algorithm Based on Ultrasonic and Monocular Vision Fusion
		4.1 Ultrasonic Ranging
		4.2 Histogram Equalization
		4.3 Image Filtering
		4.4 Edge Detection
		4.5 Obstacle Contour Detection
	5 Obstacle Avoidance Strategy
	6 Conclusion
	References
Author Index