Brain Lesion Segmentation Based on Joint Constraints of Low-Rank Representation and Sparse Representation

Abstract

The segmentation of brain lesions from a brain magnetic resonance (MR) image is of great significance for the clinical diagnosis and follow-up treatment. An automatic segmentation method for brain lesions is proposed based on the low-rank representation (LRR) and the sparse representation (SR) theory. The proposed method decomposes the brain image into the background part composed of brain tissue and the brain lesion part. Considering that each pixel in the brain tissue can be represented by the background dictionary, a low-rank representation that incorporates sparsity-inducing regularization term is adopted to model the part. Then, the linearized alternating direction method with adaptive penalty (LADMAP) was selected to solve the model, and the brain lesions can be obtained by the response of the residual matrix. The presented model not only reflects the global structure of the image but also preserves the local information of the pixels, thus improving the representation accuracy. The experimental results on the data of brain tumor patients and multiple sclerosis patients revealed that the proposed method is superior to several existing methods in terms of segmentation accuracy while realizing the segmentation automatically.

Application of Deep Learning in Civil Engineering Management

“…SRC. SRC uses sparse representation for pattern recognition problems, which characterizes unknown inputs through training samples of known types and then determines the types of test samples based on different types of reconstruction errors [13][14][15][16] , C) represents the N i training samples from the ith class. For the test sample y, the sparse representation process is as follows:…”

“…In the equation, x represents the sparse coefficient vector. At this stage, algorithms commonly used to solve sparse representation problems include the ℓ 1 norm optimization and orthogonal matching pursuit algorithm (OMP) [13][14][15][16]. According to the solution result of equation (3), i.e., 􏽢…”

“…e core of the method in this paper is to use test samples whose categories can be reliably classified and confirmed in the classification process to optimize the original classifiers, so as to obtain stronger classification performance. Specifically, the initial residual network (ResNet) [11][12][13][14] is first designed as the dominant classifier for helmet target recognition. In addition, this paper chooses the sparse representation-based classification (SRC) as the auxiliary classifier and confirms the classification of the samples to be recognized together with ResNet.…”

Deep Convolution Neural Networks Learned Image Classification for Early Cancer Detection Using Lightweight

“…However, the hybrid energy-efficient method had no effect on the error rate. [15] proposed an alternating direction method to address clinical diagnosis issues. To improve accuracy, the method was designed to reflect both the global structure of the image and the local pixel information.…”

Optimal Superpixel Kernel-Based Kernel Low-Rank and Sparsity Representation for Brain Tumour Segmentation

“…Arun and Singaravelan [13] designed a composite kernel function and applied it to the training of SVM to realise the automatic detection of brain tumours; the detection accuracy reached 93%. As classic representation learning theories, low-rank representation (LRR) and sparsity representation (SR), which mine the prior knowledge of the image by using low-rank or sparsity attributes, have been introduced into brain tumour image segmentation [14]. Due to the low-rank or sparsity constraint for the representation coefficient under the given training sample set, the structural characteristics of the image are maintained in the classification process.…”