A novel dictionary learning approach for multi-modality medical image fusion

Zhu, Zhiqin; Chai, Yi; Yin, Hongpeng; Li, Yanxia; Liu, Zhaodong

doi:10.1016/j.neucom.2016.06.036

Cited by 132 publications

(45 citation statements)

References 37 publications

(52 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…25,26 The multi-feature fusion algorithm based on SVM-D S makes full use of the superiority of SVM when solving small sample classification problems by combining the DS evidence theory in fusion of multiple feature information. It takes the feature parameters such as shape feature, texture feature, and branch dimension as the input parameters.…”

Section: Algorithm Comparison Analysismentioning

confidence: 99%

“…Thus, simply relying on the acquisition of multi-feature information to improve the recognition rate is a relatively crude method, which is only practical when certain feature information is easy to obtain. 25,26 The multi-feature fusion algorithm based on SVM-D S makes full use of the superiority of SVM when solving small sample classification problems by combining the DS evidence theory in fusion of multiple feature information. This medical recognition algorithm focuses on the relationship between feature information.…”

Section: Algorithm Comparison Analysismentioning

confidence: 99%

See 1 more Smart Citation

Fuzzy clustering recognition algorithm of medical image with multi‐resolution feature

Wang

Cui

2018

Concurrency and Computation

View full text Add to dashboard Cite

Most current medical images are collected in the presence of interference by ignored interferences such as illumination, occlusion, etc. The recognition rate is low for multi-resolution images in the case of color distortion. Therefore, a novel fuzzy clustering recognition algorithm of multi-resolution medical image was proposed in this paper. First, medical images were analyzed from both acquisition time, shooting angle, resolution, natural light measurement, and background. Secondly, in order to avoid partial occlusion, region was recalculated by correlation between Fourier and Merlin transform. Moreover, Euclidean distance between samples were identified by standardized eigenvalues, where smallest distance obtained maximum degree of membership. Experiment results showed that the proposed method had higher recognition rate (accuracy 90.46% and sensitiveness 97.89%) and stronger anti-interference than current methods. KEYWORDS fuzzy clustering, image processing, image recognition, medical image, multi-resolution 1 INTRODUCTION Today, medical image recognition is based on features of images with pattern recognition method. 1 Medical image recognition is the core content of auto-diagnosis of medical images, which is a hotspot in medical research domain. 2 Medical images contain rich features and rules of humans, which are images with high resolution, massive amount, and complex features. Research of medical image recognition has always faced challenges recently. Research and exploration of suitable algorithms for automatic recognition of medical images have important and practical significance for both theoretical and practical study. It is of great practical value to assist doctors in the clinical diagnosis by medical images. 3 In order to analyze and classify images by using intelligent algorithms, computers can be used to process medical images and extract image features such as depth information by self-learning technology. These analysis and classification improve the accuracy of medical image recognition and assist doctors in diagnosis.An image recognition method combined multiple features to achieve image classification by collecting visual information such as image color, shape, and texture. 4 Recently, there were many algorithms for medical image recognition with multi-resolution. Though some research results had been reached a good result in some.Moeskops et al proposed a medical image recognition algorithm based on multiple modalities. 5 By using cameras to obtain medical information images, this method is easy to implement, but its recognition accuracy is very low because in the analysis of the patient's medical image, there will be some areas of image features overlap, occlusion and so on. Rebouças Filho et al proposed an automatic histologically-closer classification of skin lesions. The R, G, and B components of the collected medical image are converted into H, I, and S color models for description. The use of color features to identify, simple operation, high real-time, are not sensitive to image o...

show abstract

Section: Algorithm Comparison Analysismentioning

confidence: 99%

Section: Algorithm Comparison Analysismentioning

confidence: 99%

Fuzzy clustering recognition algorithm of medical image with multi‐resolution feature

Wang

Cui

2018

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Therefore, the feedback gain K, decay rate β, parameter γ for input-to-state stability, minimal sampling interval τ min , and Lyapunov function parameter P, can be estimated by minimizing the upper bound of J. If we denote the right-hand side of inequality (19) by χ, it is:…”

Section: Theoremmentioning

confidence: 99%

Frequency Regulation of Power Systems with Self-Triggered Control under the Consideration of Communication Costs

Zhu

Sun

et al. 2017

Applied Sciences

Self Cite

View full text Add to dashboard Cite

Abstract:In control systems of power grids, conveying observations to controllers and obtaining control outputs depend greatly on communication and computation resources. Particularly for large-scale systems, the costs of computation and communication (cyber costs) should not be neglected. This paper proposes a self-triggered frequency control system for a power grid to reduce communication costs. An equation for obtaining the triggering time is derived, and an approximation method is proposed to reduce the computation cost of triggering time. In addition, the communication cost of frequency triggering is measured quantitatively and proportionally. The defined cost function considers both physical cost (electricity transmission cost) and communication cost (control signal transmission cost). The upper bound of cost is estimated. According to the estimated upper bound of cost, parameters of the controller are investigated by using the proposed optimization algorithm to guarantee the high performance of the system. Finally, the proposed self-triggered power system is simulated to verify its efficiency and effectiveness.

show abstract

“…Das et al [26] combined a non-subsampled contourlet transform (NSCT) with a reduced pulse-coupled neural network and fuzzy logic technique to overcome the image fusion problems such as contrast reduction and image degradations. Zhu et al employed a dictionary learning approach [27]. Due to the limited and redundant information in image patches created by using traditional dictionary learning methods, an alternative scheme of image patch sampling and clustering was proposed.…”

Section: Bloodmentioning

confidence: 99%

Fusion of Intraoperative 3D B-mode and Contrast-Enhanced Ultrasound Data for Automatic Identification of Residual Brain Tumors

et al. 2017

View full text Add to dashboard Cite

Intraoperative ultrasound (iUS) imaging is routinely performed to assist neurosurgeons during tumor surgery. In particular, the identification of the possible presence of residual tumors at the end of the intervention is crucial for the operation outcome. B-mode ultrasound remains the standard modality because it depicts brain structures well. However, tumorous tissue is hard to differentiate from resection cavity borders, blood and artifacts. On the other hand, contrast enhanced ultrasound (CEUS) highlights residuals of the tumor, but the interpretation of the image is complex. Therefore, an assistance system to support the identification of tumor remnants in the iUS data is needed. Our approach is based on image segmentation and data fusion techniques. It consists of combining relevant information, automatically extracted from both intraoperative B-mode and CEUS image data, according to decision rules that model the analysis process of neurosurgeons to interpret the iUS data. The method was tested on an image dataset of 23 patients suffering from glioblastoma. The detection rate of brain areas with tumor residuals reached by the algorithm was qualitatively and quantitatively compared with manual annotations provided by experts. The results showed that the assistance tool was able to successfully identify areas with suspicious tissue.

show abstract

A novel dictionary learning approach for multi-modality medical image fusion

Cited by 132 publications

References 37 publications

Fuzzy clustering recognition algorithm of medical image with multi‐resolution feature

Fuzzy clustering recognition algorithm of medical image with multi‐resolution feature

Frequency Regulation of Power Systems with Self-Triggered Control under the Consideration of Communication Costs

Fusion of Intraoperative 3D B-mode and Contrast-Enhanced Ultrasound Data for Automatic Identification of Residual Brain Tumors

Contact Info

Product

Resources

About