14 Medical Image Fusion Method by Deep Learning
-
R. Sabitha
Abstract
Scientific image fusion is a method that mixes a couple of clinical snap shots from exclusive modalities to generate a single composite picture with improved diagnostic information. Traditional image fusion strategies rely upon handmade capabilities and predefined rules, which won’t fully exploit the rich information contained in medical pictures. In recent years, deep learning processes have shown great potential in numerous clinical image processing tasks. In this chapter, we propose a novel scientific image fusion approach using deep learning. Our approach uses a convolutional neural network to learn features from the input pix and then fuses them at a later stage using a fusion layer. Unlike traditional techniques, our approach automatically learns the optimal fusion rules from the input images, thus reducing the need for manual intervention. We also introduce a new dataset including multimodal medical snap shots for evaluation. Experiments on this dataset demonstrate the effectiveness of our technique in generating fused images with improved information content compared to traditional methods. Our technique also outperforms current deep learning- based fusion strategies on this dataset. Furthermore, we demonstrate the robustness of our approach by testing on images with varying levels of noise and artifacts. Overall, our technique shows promise in enhancing the diagnostic potential of clinical pixels and has potential applications in disease diagnosis and monitoring.
Abstract
Scientific image fusion is a method that mixes a couple of clinical snap shots from exclusive modalities to generate a single composite picture with improved diagnostic information. Traditional image fusion strategies rely upon handmade capabilities and predefined rules, which won’t fully exploit the rich information contained in medical pictures. In recent years, deep learning processes have shown great potential in numerous clinical image processing tasks. In this chapter, we propose a novel scientific image fusion approach using deep learning. Our approach uses a convolutional neural network to learn features from the input pix and then fuses them at a later stage using a fusion layer. Unlike traditional techniques, our approach automatically learns the optimal fusion rules from the input images, thus reducing the need for manual intervention. We also introduce a new dataset including multimodal medical snap shots for evaluation. Experiments on this dataset demonstrate the effectiveness of our technique in generating fused images with improved information content compared to traditional methods. Our technique also outperforms current deep learning- based fusion strategies on this dataset. Furthermore, we demonstrate the robustness of our approach by testing on images with varying levels of noise and artifacts. Overall, our technique shows promise in enhancing the diagnostic potential of clinical pixels and has potential applications in disease diagnosis and monitoring.
Chapters in this book
- Frontmatter I
- Contents V
- List of Authors IX
- 1 Medical Image Processing: A Multimodal Fusion Technique 1
- 2 Image Fusion Mathematics Theory and Practice 21
- 3 Current Trends in High-Resolution Image Reconstruction 43
- 4 Image Fusion Through Multiresolution Oversampled Decompositions 59
- 5 Mathematical Models for Remote Sensing Image Processing 75
- 6 Component Analysis and Medical Image Fusion 91
- 7 Soft Computing Approaches to Medical Image Fusion 105
- 8 Mathematical Techniques in Multispectral Image Fusion 119
- 9 Fusion of Artificial Intelligence and Machine Learning for Advanced Image Processing 135
- 10 Fusion of Multispectral and Panchromatic Images as an Optimization Problem 151
- 11 Image Fusion Using Optimization of Statistical Measurements 167
- 12 Empirical Mode Decomposition for Simultaneous Image Enhancement and Fusion 183
- 13 Multimodality Sensor Image Fusion 189
- 14 Medical Image Fusion Method by Deep Learning 207
- 15 Quaternion-Based Sparse Techniques for Multimodal and Multispectral Image Processing 223
- 16 Quaternion Neural Networks for Geometrical Operators in High-Dimensional Quaternion Space 239
- 17 Image Dehazing Using Quaternion Complex Algebra-Based Neural Networks 255
- 18 Deep Learning Model for Image Fusion and Accurate Classification of Remote Sensing Images 271
- 19 Multimodal Medical Supervised Image Fusion Method 297
- 20 Medical Image Fusion Using Deep Learning Mechanisms 315
- 21 Multifocus Image Fusion Using Content-Adaptive Blurring 339
- 22 Optimizing Top Precision Performance Measure of Content-Based Image Retrieval by Learning Similarity Function 349
- 23 Performance Evaluation of Image Fusion Techniques 361
- Index 373
Chapters in this book
- Frontmatter I
- Contents V
- List of Authors IX
- 1 Medical Image Processing: A Multimodal Fusion Technique 1
- 2 Image Fusion Mathematics Theory and Practice 21
- 3 Current Trends in High-Resolution Image Reconstruction 43
- 4 Image Fusion Through Multiresolution Oversampled Decompositions 59
- 5 Mathematical Models for Remote Sensing Image Processing 75
- 6 Component Analysis and Medical Image Fusion 91
- 7 Soft Computing Approaches to Medical Image Fusion 105
- 8 Mathematical Techniques in Multispectral Image Fusion 119
- 9 Fusion of Artificial Intelligence and Machine Learning for Advanced Image Processing 135
- 10 Fusion of Multispectral and Panchromatic Images as an Optimization Problem 151
- 11 Image Fusion Using Optimization of Statistical Measurements 167
- 12 Empirical Mode Decomposition for Simultaneous Image Enhancement and Fusion 183
- 13 Multimodality Sensor Image Fusion 189
- 14 Medical Image Fusion Method by Deep Learning 207
- 15 Quaternion-Based Sparse Techniques for Multimodal and Multispectral Image Processing 223
- 16 Quaternion Neural Networks for Geometrical Operators in High-Dimensional Quaternion Space 239
- 17 Image Dehazing Using Quaternion Complex Algebra-Based Neural Networks 255
- 18 Deep Learning Model for Image Fusion and Accurate Classification of Remote Sensing Images 271
- 19 Multimodal Medical Supervised Image Fusion Method 297
- 20 Medical Image Fusion Using Deep Learning Mechanisms 315
- 21 Multifocus Image Fusion Using Content-Adaptive Blurring 339
- 22 Optimizing Top Precision Performance Measure of Content-Based Image Retrieval by Learning Similarity Function 349
- 23 Performance Evaluation of Image Fusion Techniques 361
- Index 373
