A new robust method for blood vessel segmentation in retinal fundus images based on weighted line detector and hidden Markov model. (April 2020)

• Record Type: Journal Article
• Authors: Zhou, Chao; Zhang, Xiaogang; Chen, Hua
• Is Part Of: Computer methods and programs in biomedicine. Volume 187(2020)
• Publication Date: 2020-04
• Subjects: Retinal image -- Difficult region -- Thin vessel -- Vessel segmentation -- Line detector -- Hidden Markov model; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2019.105231
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: In this paper, three new algorithms are proposed to deal with different problems in retinal vessel segmentation. Thus, the experiment results show that our method is effective in dealing with thin vessel missing and eliminating wrong segmentation of some difficult regions which is also reflected by the high SP score achieved by our algorithms. Overall, our method achieves a high segmentation accuracy which is comparable to these of published state-of-the-art methods and better than the result of a human observer. According to new SSIM and FSIM measures, our result has an excellent similarity to ground truth. The proposed algorithms are designed to segment retinal vessels, but they have other potential applications. For example, the new line detector is a very fast unsupervised approach to detect wirelike structure. Thus, it can be easily extended to perform on high resolution retinal images or used to abstract other wirelike structures such as plant root [41] or neuron axon [42] . The HMM-based tracing approach is presented to detect the vessel centerlines and thus can potentially contribute to other applications in retinal fundus image such as retinal vessel landmarks [43], that is, distinguishing vessel bifurcations and crossings. Abstract: Background and objective: Automatic vessel segmentation is a crucial preliminary processing step to facilitate ophthalmologist diagnosis in some diseases. But, due to the complexity of retinal fundus image, there are some problems on accurate segmentation of retinal vessel. In this paper, a new method for retinal vessel segmentation is proposed to handle two main problems: thin vessel missing and false detection in difficult regions. Methods: First, an improved line detector is proposed and used to fast extract the major structures of vessels. Then, Hidden Markov model (HMM) is applied to effectively detect vessel centerlines that include thin vessels. Finally, a denoising approach is presented to remove noises and two types of vessels are unified to obtain the complete segmentation results. Results: Our method is tested on two public databases (DRIVE and STARE databases), and five measures namely accuracy (Acc), sensitivity (Se), specificity (Sp), Dice coefficient (Dc), structural similarity index (SSIM) and feature similarity index (FSIM) are used to evaluate our segmentation performance. The respective values of the performance measures are 0.9475, 0.7262, 0.9803, 0.7781, 0.9992 and 0.9793 for DRIVE dataset and 0.9535, 0.7865, 0.9730, 0.7764, 0.9987 and 0.9742 for STARE dataset. Conclusions: The experiment results show that our method outperforms most published state-of-the-art methods and is better the result of a human observer. Moreover, in term of specificity, our proposed algorithm can obtain the best score among the unsupervised methods. Meanwhile, there are excellent structure and feature similarities between our result and the ground truth according to achieved SSIM and FSIM. Visual inspection on the segmentation results shows that the proposed method produces more accurate segmentations on some difficult regions such as optic disc and central light reflex while detecting thin vessels effectively compared with the other methods.