Chengdong Wu

Bio: Chengdong Wu is an academic researcher from Northeastern University. The author has contributed to research in topics: Optic disc & Active contour model. The author has an hindex of 3, co-authored 4 publications receiving 23 citations.

Optic Disc and Cup Segmentation in Retinal Images for Glaucoma Diagnosis by Locally Statistical Active Contour Model with Structure Prior.

Abstract: Accurate optic disc and optic cup segmentation plays an important role for diagnosing glaucoma. However, most existing segmentation approaches suffer from the following limitations. On the one hand, image devices or illumination variations always lead to intensity inhomogeneity in the fundus image. On the other hand, the spatial prior knowledge of optic disc and optic cup, e.g., the optic cup is always contained inside the optic disc region, is ignored. Therefore, the effectiveness of segmentation approaches is greatly reduced. Different from most previous approaches, we present a novel locally statistical active contour model with the structure prior (LSACM-SP) approach to jointly and robustly segment the optic disc and optic cup structures. First, some preprocessing techniques are used to automatically extract initial contour of object. Then, we introduce the locally statistical active contour model (LSACM) to optic disc and optic cup segmentation in the presence of intensity inhomogeneity. Finally, taking the specific morphology of optic disc and optic cup into consideration, a novel structure prior is proposed to guide the model to generate accurate segmentation results. Experimental results demonstrate the advantage and superiority of our approach on two publicly available databases, i.e., DRISHTI-GS and RIM-ONE r2, by comparing with some well-known algorithms.

Brain tumor segmentation in MR images using a sparse constrained level set algorithm

Abstract: Brain tumor segmentation using Magnetic Resonance (MR) Imaging technology plays a significant role in computer-aided brain tumor diagnosis. However, when applying classic segmentation methods, limitations such as inhomogeneous intensity, complex physiological structure and blurred tissues boundaries in brain MR images usually lead to unsatisfactory results. To address these issues, this paper proposes an automatic sparse constrained level set method to realize the brain tumor segmentation in MR images. By studying brain tumor images, this method finds out common characteristics of brain tumors’ shape and constructs a sparse representation model. By considering this model as a prior constraint, an energy function based on level set method is constructed. In experiments, the proposed method can achieve an average accuracy of 96.20% for the MR images from the dataset Brats2017 and performs better than the others. With lower false positive rate and stronger robustness, the experimental results show that the proposed method can segment brain tumor from MR image accurately and stably.

Accurate and Efficient Segmentation of Optic Disc and Optic Cup in Retinal Images Integrating Multi-View Information

Abstract: Glaucoma is an eye disease which is one of the most common causes of blindness. Accurate optic disc (OD) and optic cup (OC) segmentation play a critical role for detecting glaucoma. Considering that the existing approaches can’t effectively integrate the multi-view information deriving from shape and appearance to sufficiently describe OD and OC for segmentation, Locally Statistical Active Contour Model with the Information of Appearance and Shape (LSACM-AS) and Modified Locally Statistical Active Contour Model with the Information of Appearance and Shape (MLSACM-AS) are proposed in this paper. The main contributions are as below: (1) we introduce the Locally Statistical Active Contour Model (LSACM) to address the commonly occurred intensity inhomogeneity phenomenon caused by imperfection of image devices or illumination variations. (2) In order to overcome the common effects caused by pathological changes (i.e., peripapillary atrophy (PPA)) and vessel occlusion in OD and OC segmentation, we integrate the local image probability information around the point of interest from a multi-dimensional feature space into our model to preserve the integrity of the OD and OC structures. (3) Since the segmentation objects have the similar ellipse shape structure, we incorporate the shape priori constraint information into our model to further improve the robustness of the variations found in and around objects regions. To evaluate the effectiveness of the proposed models, an available publicly DRISHTI-GS database is employed in this paper. Seen from the abundant experiments, the proposed models outperform the state-of-the-art approaches in terms of the obtained qualitative and quantitative results.

Automatic optic disc detection using low-rank representation based semi-supervised extreme learning machine

Abstract: Optic disc detection plays an important role in developing automatic screening systems for diabetic retinopathy. Several supervised learning-based approaches have been proposed for optic disc detection. However, these approaches demand that the input training examples are completely labelled. Essentially, in medical image analysis, it is difficult to prepare several training samples which were given reliable class labels due to the fact that manually labelling data is very expensive. Moreover, retinal images such as complex vessels structures in the optic disc constituting nonlinear relationships in high-dimensional observation space, which cannot work well by traditional linear classifiers. In this study, a novel approach named low-rank representation based semi-supervised extreme learning machine (LRR-SSELM) is proposed for automated optic disc detection. Our model has the following advantages. First, it detects the optic disc from the viewpoint of semi-supervised learning and overcomes the problem there are small portion of labelled samples. Second, a nonlinear classifier is introduced into our model to fully explore the nonlinear data. Third, the local and global structures of original data can be greatly persevered by low-rank representation (LRR). The performance of the proposed method is validated on three publicly available databases, DIARETDB0, DIARETDB1 and Messidor. The experimental results indicate the advantages and effectiveness of the proposed approach.