Medical image alignment based on landmark- and approximate contour-matching

With the development of imaging-guided surgery and radiotherapy, the clinical demand for medical image matching research is stronger and the challenges are even greater. In recent years, deep learning, especially deep convolution neural networks, has made excellent achievements in medical image processing, and the research on medical image matching has developed rapidly. In this paper, the domestic and foreign research progress based on deep learning medical image alignment is summarized according to the classification of technical methods, including the similarity estimation based on optimization strategy, the transformation parameters of direct estimation of medical image alignment, etc. Then it analyses the challenge of the deep learning method in medical image matching and puts forward possible solutions and research directions. Key words: medical image recording, deep learning, CNN, full convolutional network

Download Full-text

PERFORMANCE COMPARISON OF THE POWELL AND PSO METHODS FOR MEDICAL IMAGE ALIGNMENT

Biomedical Engineering Applications Basis and Communications ◽

10.4015/s1016237217500144 ◽

2017 ◽

Vol 29 (02) ◽

pp. 1750014

Author(s):

Meisen Pan ◽

Fen Zhang

Keyword(s):

Particle Swarm Optimization ◽

Medical Image ◽

Optimization Methods ◽

Optimization Method ◽

Performance Comparison ◽

Image Alignment ◽

Swarm Optimization ◽

Similarity Metric ◽

Transformation Parameters ◽

Powell Method

In the course of aligning medical images, the similarity metric (also called alignment function) is regarded as the objective function, and the optimization method as the tool for exploring the optimal transformation parameters. In this paper, the medical image alignment is represented first and then the optimization methods are depicted in detail. With this description, by the use of the mutual information (MI) as the similarity metric, the Powell method and the particle swarm optimization (PSO) method are employed to seek the optimal transformation parameters respectively, and their optimization performances are estimated and compared. The experimental results show that the Powell and PSO methods can cater to both the multi-modality medical image alignments.

Download Full-text

Unsupervised Medical Image Alignment With Curriculum Learning

10.1109/icip42928.2021.9506067 ◽

2021 ◽

Author(s):

Mihail Burduja ◽

Radu Tudor Ionescu

Keyword(s):

Medical Image ◽

Image Alignment

Download Full-text

Medical Image Alignment by Normal Vector Information

Computational Intelligence and Security - Lecture Notes in Computer Science ◽

10.1007/11596448_132 ◽

2005 ◽

pp. 890-895

Author(s):

Xiahai Zhuang ◽

Lixu Gu ◽

Jianfeng Xu

Keyword(s):

Medical Image ◽

Image Alignment ◽

Normal Vector ◽

Vector Information

Download Full-text

An overlap invariant entropy measure of 3D medical image alignment

Pattern Recognition ◽

10.1016/s0031-3203(98)00091-0 ◽

1999 ◽

Vol 32 (1) ◽

pp. 71-86 ◽

Cited By ~ 1410

Author(s):

C. Studholme ◽

D.L.G. Hill ◽

D.J. Hawkes

Keyword(s):

Medical Image ◽

Image Alignment ◽

Entropy Measure

Download Full-text

A Comparative Study of Ant and Genetic Algorithms in Digital Mammography

International Journal of Scientific Research in Computer Science Engineering and Information Technology ◽

10.32628/cseit183863 ◽

2018 ◽

pp. 194-200

Author(s):

J. Magelin Mary ◽

Chitra K. ◽

Y. Arockia Suganthi

Keyword(s):

Genetic Algorithm ◽

Image Processing ◽

Medical Image ◽

Medical Image Processing ◽

Processing Technique ◽

Input Image ◽

Image Processing Technique ◽

Computational Time ◽

Image Mining ◽

The Individual

Image processing technique in general, involves the application of signal processing on the input image for isolating the individual color plane of an image. It plays an important role in the image analysis and computer version. This paper compares the efficiency of two approaches in the area of finding breast cancer in medical image processing. The fundamental target is to apply an image mining in the area of medical image handling utilizing grouping guideline created by genetic algorithm. The parameter using extracted border, the border pixels are considered as population strings to genetic algorithm and Ant Colony Optimization, to find out the optimum value from the border pixels. We likewise look at cost of ACO and GA also, endeavors to discover which one gives the better solution to identify an affected area in medical image based on computational time.

Download Full-text