scholarly journals Discontinuity Preserving Image Registration through Motion Segmentation: A Primal-Dual Approach

2016 ◽  
Vol 2016 ◽  
pp. 1-20 ◽  
Author(s):  
Silja Kiriyanthan ◽  
Ketut Fundana ◽  
Tahir Majeed ◽  
Philippe C. Cattin
Keyword(s):  
Image Registration ◽  
Medical Image ◽  
Motion Segmentation ◽  
Medical Image Analysis ◽  
Organ Motion ◽  
Mr Images ◽  
Clinical Routine ◽  
Dual Approach ◽  
Sliding Motion ◽  
Primal Dual

Image registration is a powerful tool in medical image analysis and facilitates the clinical routine in several aspects. There are many well established elastic registration methods, but none of them can so far preserve discontinuities in the displacement field. These discontinuities appear in particular at organ boundaries during the breathing induced organ motion. In this paper, we exploit the fact that motion segmentation could play a guiding role during discontinuity preserving registration. The motion segmentation is embedded in a continuous cut framework guaranteeing convexity for motion segmentation. Furthermore we show that a primal-dual method can be used to estimate a solution to this challenging variational problem. Experimental results are presented for MR images with apparent breathing induced sliding motion of the liver along the abdominal wall.

An Automatic Medical Image Registration Approach Based on Common Sub-regions of Interest

2019 ◽  
Vol 9 (2) ◽  
pp. 251-260
Author(s):  
Fakhre Alam ◽  
Sami UR Rahman ◽  
Nasser Tairan ◽  
Habib Shah ◽  
Mohammed Saeed Abohashrh ◽  
...  
Keyword(s):  
Image Registration ◽  
Medical Image ◽  
Medical Images ◽  
Medical Image Analysis ◽  
Source Image ◽  
Medical Procedures ◽  
Mr Images ◽  
Time Efficiency ◽  
Guided Surgery ◽  
Global Registration

Accurate and efficient image registration, based on interested common sub-regions is still a challenging task in medical image analysis. This paper presents an automatic features based approach for the rigid and deformable registration of medical images using interested common sub-regions. In the proposed approach, interested common sub-regions in two images (target image and source image) are automatically detected and locally registered. The final global registration is performed, using the transformation parameters obtained from the local registration. Registration using interested common sub-regions is always required in image guided surgery (IGS) and other medical procedures because it considers only the desired objects in medical images instead of the whole image contents. The proposed interested common sub-regions based registration is compared with the two states-of-the-art methods on MR images of human brain. In the experiments of rigid and deformable registrations, we show that our approach outperforms in terms of both the accuracy and time efficiency. The results reveal that interested common sub-region based registration can achieve good performance, regarding both the accuracy as well as the the time efficiency in monomodal brain image registration. In addition, the proposed approach also indicates the potential for multimodal images of different human organs.

Medical Image Analysis by Robot Kinematics

2006 ◽  
Vol 326-328 ◽  
pp. 875-878
Author(s):  
Jae Bum An ◽  
Li Li Xin
Keyword(s):  
Image Registration ◽  
Medical Image ◽  
Medical Images ◽  
Medical Image Analysis ◽  
Robot Kinematics ◽  
Surgical Robots ◽  
Cross Sectional ◽  
Registration Algorithm ◽  
Straight Line ◽  
Actual Shape

In this paper we present an analysis of medical images based on robot kinematics. One of the most important problems in robot-assisted surgeries is associated with the medical image registration of surgical tools and anatomical targets. The fundamental problems of contemporary frame-based image registration are that the registration fails in case of incomplete data in the image and the registration algorithm depends on the shape, assembly, and number of fiducials. To solve the registration problem in the situation where a cylindrical end-effector of surgical robots operates inside the patient’s body, we developed a numerical method by applying robot kinematics knowledge to cross-sectional medical images. Our method includes a 6-D registration algorithm and a cylindrical frame with four helix and one straight line fiducials. The numerical algorithm requires only a single cross-sectional image and are robust to noise and missing data, and are algorithmically invariant to the actual shape, number, and assembly of fiducials. The algorithm and frame are introduced in this paper, and simulation results are described to show the adequate accuracy and resistance to noise.

FAST, MULTI-MODAL AND DISCONTINUITY-PRESERVING IMAGE REGISTRATION USING MUTUAL INFORMATION

2013 ◽  
Vol 22 (06) ◽  
pp. 1360015
Author(s):  
GIORGIO PANIN
Keyword(s):  
Image Registration ◽  
Mutual Information ◽  
Parallel Implementation ◽  
Medical Image Analysis ◽  
Hessian Matrix ◽  
Stereo Image ◽  
Graphics Hardware ◽  
Ground Truth Data ◽  
Primal Dual ◽  
Image Pairs

In this paper, we describe a fast and efficient method for multi-modal and discontinuity-preserving image registration, implemented on graphics hardware. Multi-sensory data fusion and medical image analysis often pose the challenging task of aligning dense, non-rigid and multi-modal images. However, also optical sequences or stereo image pairs may present variable illumination conditions and noise. The above problems can be addressed by an invariant similarity measure, such as mutual information. Additionally, when using a regularized approach to deal with the ill-posedness of the problem, one has to take care of preserving discontinuities at the motion boundaries. Our approach efficiently addresses the above issues through a primal-dual convex estimation framework, using an approximated Hessian matrix that decouples pixel dependencies, while being asymptotically correct. At the same time, we achieve a high computational efficiency by means of pre-quantized kernel density estimation and differentiation, as well as a parallel implementation on the GPU. Our approach is demonstrated on ground-truth data from the Middlebury database, as well as medical and visible-infrared image pairs.

An Improved Demon Registration with Mutual Information for Non-Rigid Medical Images

2013 ◽  
Vol 284-287 ◽  
pp. 1622-1626 ◽  
Author(s):  
Xiao Hui Xie ◽  
Cui Ma ◽  
Qiang Sun ◽  
Ru Xu Du
Keyword(s):  
Image Analysis ◽  
Image Registration ◽  
Mutual Information ◽  
Critical Points ◽  
Medical Image ◽  
Medical Image Analysis ◽  
Good Method ◽  
Demons Algorithm ◽  
Single Modality ◽  
Ct And Mri

Non-rigid image registration plays an important role in medical imaging. Classic Demons algorithm is a good method for image registration in some domain. One disadvantage of classic Demons algorithm is that the topological preservation can not be ensured, and it can only adapt to deal with the single modality image registration. In medical image analysis, the different modal images comparison and fusion are needed to give the doctor enough information for making a decision. The mutual information algorithm has been validated useful for multi-modality image registration. By analyzing the critical points of Demons registration like mis-registration, an improved Demons algorithm with mutual information evaluation is proposed. Experiment results on liver images between CT and MRI modality show that the proposed algorithm can deal with multi-modality image registration well and it can hold the abilities even faces the noise and distortion.

scholarly journals Comparative Analysis and Detection of Brain Tumor Using Fusion Technique of T1 and T2 Weighted MR Images

2021 ◽  
Vol 11 (1) ◽  
pp. 54-61
Author(s):  
Padmanjali A. Hagargi
Keyword(s):  
Brain Tumor ◽  
Image Fusion ◽  
Medical Image ◽  
Medical Image Analysis ◽  
Single Image ◽  
Mr Images ◽  
Fusion Technique ◽  
Fused Image ◽  
Proper Diagnosis ◽  
Image Fusion Technique

Image fusion is a technique to fuse the two or more images. As the fused image gathers more information as comparative to the single image, image fusion of multiple images can be done to extract more number of information, with this reason the it is important in the field of medical image analysis. The fusion technique is so useful in detection of different kind of disease using different kind of medical images. Brain tumor disease is a large issue because of non-proper diagnosis and treatment is lacking accordingly. Using T1, T2 Weighted MR images are two medical MR images at different time constant during the scanning of brain tumor. These two or more images can be used to extract more information by the various image fusion technique.

scholarly journals A Medical Image Registration Method Based on Progressive Images

2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Qian Zheng ◽  
Qiang Wang ◽  
Xiaojuan Ba ◽  
Shan Liu ◽  
Jiaofen Nan ◽  
...  
Keyword(s):  
Image Registration ◽  
Mutual Information ◽  
Medical Image ◽  
Medical Image Analysis ◽  
Reference Image ◽  
Medical Image Registration ◽  
Registration Method ◽  
Multimodal Registration ◽  
Coarse Registration ◽  
Registration Result

Background. Medical image registration is an essential task for medical image analysis in various applications. In this work, we develop a coarse-to-fine medical image registration method based on progressive images and SURF algorithm (PI-SURF) for higher registration accuracy. Methods. As a first step, the reference image and the floating image are fused to generate multiple progressive images. Thereafter, the floating image and progressive image are registered to get the coarse registration result based on the SURF algorithm. For further improvement, the coarse registration result and the reference image are registered to perform fine image registration. The appropriate progressive image has been investigated by experiments. The mutual information (MI), normal mutual information (NMI), normalized correlation coefficient (NCC), and mean square difference (MSD) similarity metrics are used to demonstrate the potential of the PI-SURF method. Results. For the unimodal and multimodal registration, the PI-SURF method achieves the best results compared with the mutual information method, Demons method, Demons+B-spline method, and SURF method. The MI, NMI, and NCC of PI-SURF are improved by 15.5%, 1.31%, and 7.3%, respectively, while MSD decreased by 13.2% for the multimodal registration compared with the optimal result of the state-of-the-art methods. Conclusions. The extensive experiments show that the proposed PI-SURF method achieves higher quality of registration.

Diagnosing Alzheimer’s Disease based on Multiclass MRI Scans using Transfer Learning Techniques

2021 ◽  
Vol 17 ◽  
Author(s):  
Deekshitha Prakash ◽  
Nuwan Madusanka ◽  
Subrata Bhattacharjee ◽  
Cho-Hee Kim ◽  
Hyeon-Gyun Park ◽  
...  
Keyword(s):  
Neural Network ◽  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Image Analysis ◽  
Magnetic Resonance ◽  
Transfer Learning ◽  
Medical Image ◽  
Medical Image Analysis ◽  
Mr Images ◽  
Learning Technique

Background: In this study, we employed transfer learning technique to classify Magnetic Resonance (MR) images using a pre-trained convolutional neural network (CNN). Aims: To prevent Alzheimer’s disease (AD) from progression to dementia, early prediction and classification of AD plays a crucial role in medical image analysis. Background: In this study, we employed transfer learning technique to classify Magnetic Resonance (MR) images using a pre-trained convolutional neural network (CNN). Objective: To address the early diagnosis of AD, we employed computer-assisted technique specifically deep learning (DL) model to detect AD. Methods: In particular, we classified Alzheimer’s disease (AD), mild cognitive impairment (MCI) and normal control (NC) subjects using whole slide two-dimensional (2D) images. To illustrate this approach, we made use of state-of-the-art CNN base models, i.e., the residual networks ResNet-101, ResNet-50 and ResNet-18, and compared their effectiveness to identifying AD. To evaluate this approach, an AD Neuroimaging Initiative (ADNI) dataset was utilized. We have also showed uniqueness by using MR images selected only from the central slice containing left and right hippocampus regions to evaluate the models. Results: All the three models used randomly split data in the ratio 70:30 for training and testing. Among the three, ResNet-101 showed 98.37% accuracy, better than the other two ResNet models, and performed well in multiclass classification. The promising results emphasize the benefit of using transfer learning specifically when the dataset is low. Conclusion: From this study, we can assure that transfer learning helps to overcome DL problems mainly when the data available is insufficient to train a model from scratch. This approach is highly advantageous in medical image analysis to diagnose diseases like AD.

Privacy-preserving Collaborative Training for Medical Image Analysis Based on Multi-Blockchain

2020 ◽  
Vol 23 ◽  
Author(s):  
Wanlu Zhang ◽  
Qigang Wang ◽  
Mei Li
Keyword(s):  
Medical Image ◽  
Data Privacy ◽  
Medical Image Analysis ◽  
Auxiliary Information ◽  
Training Process ◽  
Private Data ◽  
Medical Institutions ◽  
Model Training ◽  
Collaborative Training ◽  
Similar Task

Background: As artificial intelligence and big data analysis develop rapidly, data privacy, especially patient medical data privacy, is getting more and more attention. Objective: To strengthen the protection of private data while ensuring the model training process, this article introduces a multi-Blockchain-based decentralized collaborative machine learning training method for medical image analysis. In this way, researchers from different medical institutions are able to collaborate to train models without exchanging sensitive patient data. Method: Partial parameter update method is applied to prevent indirect privacy leakage during model propagation. With the peer-to-peer communication in the multi-Blockchain system, a machine learning task can leverage auxiliary information from another similar task in another Blockchain. In addition, after the collaborative training process, personalized models of different medical institutions will be trained. Results: The experimental results show that our method achieves similar performance with the centralized model-training method by collecting data sets of all participants and prevents private data leakage at the same time. Transferring auxiliary information from similar task on another Blockchain has also been proven to effectively accelerate model convergence and improve model accuracy, especially in the scenario of absence of data. Personalization training process further improves model performance. Conclusion: Our approach can effectively help researchers from different organizations to achieve collaborative training without disclosing their private data.

Sign in / Sign up

Export Citation Format

Share Document