290 Hits in 6.2 sec

U-Net with Graph Based Smoothing Regularizer for Small Vessel Segmentation on Fundus Image [article]

Lukman Hakim, Novanto Yudistira, Muthusubash Kavitha, Takio Kurita
2020 arXiv   pre-print
Paying attention to the low contrast small blood vessel in fundus region, first time we proposed to combine graph based smoothing regularizer with the loss function in the U-net framework.  ...  The potential of the proposed graph based smoothing regularizer in reconstructing small vessel is compared over the classical U-net with or without regularizer.  ...  Conclusions We newly proposed a graph based smoothing regularization term with the loss function in the U-net framework for the segmentation of small vessels in the retinal image.  ... 
arXiv:2009.07567v1 fatcat:72ty4hmehrh7nk2lijnynpan7e

Deep Guidance Network for Biomedical Image Segmentation

Pengshuai Yin, Rui Yuan, Yiming Cheng, Qingyao Wu
2020 IEEE Access  
Segmentation of 2D images is a fundamental problem for biomedical image analysis. The most widely used architecture for biomedical image segmentation is U-Net.  ...  Our method enables end to end training and fast inference (43ms for one image). We conduct extensive experiments for the task of vessel segmentation and optic disc and cup segmentation.  ...  Early attempts for OC and OD segmentation in fundus images are based on hand-craft features such as color [27] , contrast thresholding [28] , level set approach [29] , clustering based methods [30]  ... 
doi:10.1109/access.2020.3002835 fatcat:rv66rxz5dfe3to6pcmqqqvp3bq

Improvement for Single Image Super-resolution and Image Segmentation by Graph Laplacian Regularizer Based on Differences of Neighboring Pixels

2022 International Journal of Intelligent Engineering and Systems  
This study presents a regularization term based on the differences of neighboring pixels. We define the differences of neighboring pixels by using the Graph Theory approach.  ...  Qualitative and quantitative results show that our developed regularizer proved to enhance the boundary structure on image super-resolution and image segmentation tasks, which is achieved Structural Similarity  ...  U-Net U-Net + GLRDN 0.7625 0.8040 0.9624 0.9326 0.9497 0.9580 0.9350 0.9561 Figure. 2 Comparison results of blood vessel segmentation between our proposed regularizer and without proposed regularizer  ... 
doi:10.22266/ijies2022.0228.10 fatcat:fyaondf2ojcxjncyqswcoyfbmm

Fully Automated Tree Topology Estimation and Artery-Vein Classification [article]

Aashis Khanal, Saeid Motevali, Rolando Estrada
2022 arXiv   pre-print
We present a fully automatic technique for extracting the retinal vascular topology, i.e., how the different vessels are connected to each other, given a single color fundus image.  ...  As our experiments show, our topology-based artery-vein labeling achieved state-of-the-art results on multiple datasets.  ...  U-Net, a multi-level encoder-decoder based architecture is uses as a backbone for major medical image segmentation task [16] .  ... 
arXiv:2202.02382v1 fatcat:daqpguvdknaqdh2xgl34nm6bri

Retinal Vessel Segmentation Using Deep Learning: A Review

Chunhui Chen, Joon Huang Chuah, Ali Raza, Yizhou Wang
2021 IEEE Access  
In this work, we reviewed recent publications for retinal vessel segmentation based on deep learning.  ...  INDEX TERMS Retinal vessel segmentation, fundus images, deep learning, convolutional neural network.  ...  Compared with traditional CNNs, FCNs can predict each pixel in an image or image patch, so it is more suitable and fast for image segmentation tasks. C. U-NET Ronneberger, et al.  ... 
doi:10.1109/access.2021.3102176 fatcat:x3jqvx67qndgnoi7wi357b6zwe

Coronary Vessel Segmentation by Coarse-to-Fine Strategy Using U-nets

Le Nhi Lam Thuy, Tan Dat Trinh, Le Hoang Anh, Jin Young Kim, Huynh Trung Hieu, Pham The Bao, Changming Sun
2021 BioMed Research International  
Using these junctions, we determine regions of the secondary coronary arteries (rectangular regions) for a secondary coronary artery-extracted segment with the second U-net model.  ...  To overcome this problem, we propose a novel segmenting method for coronary artery extraction from angiograms based on the primary and secondary coronary artery.  ...  frameworks for coronary vessel segmentation and an extended method to deal with 3D images.  ... 
doi:10.1155/2021/5548517 pmid:33898624 pmcid:PMC8052146 fatcat:jrynubgyojgqlc7trdbhjwnapq

Automatic Segmentation and Visualization of Choroid in OCT with Knowledge Infused Deep Learning [article]

Huihong Zhang, Jianlong Yang, Kang Zhou, Fei Li, Yan Hu, Yitian Zhao, Ce Zheng, Xiulan Zhang, Jiang Liu
2020 IEEE journal of biomedical and health informatics   accepted
To eliminate the retinal vessel shadows, we propose a pipeline that firstly use anatomical and OCT imaging knowledge to locate the shadows using their projection on the retinal pigment epithelium layer  ...  In this paper, we propose to incorporate medical and imaging prior knowledge with deep learning to address these two problems.  ...  ACKNOWLEDGEMENT We would like to thank all the reviewers for their insightful comments and concerns which stimulate us to improve our work.  ... 
doi:10.1109/jbhi.2020.3023144 pmid:32931435 arXiv:2002.04712v2 fatcat:nfcxejpqy5dbdora35prdjobee

Comparative Analysis of Vessel Segmentation Techniques in Retinal Images

Azhar Imran, Jianqiang Li, Yan Pei, Ji-Jiang Yang, Qing Wang
2019 IEEE Access  
INDEX TERMS Vessel segmentation, retinal diseases, image segmentation, retinal fundus images, medical imaging. 114866 VOLUME 7, 2019 A.  ...  This survey presents a comparative analysis of various machine learning and deep learning-based methods for automated blood vessel segmentation in retinal images.  ...  [28] proposed a Dense U-Net based on patch-based learning method for retinal vessel segmentation.  ... 
doi:10.1109/access.2019.2935912 fatcat:64ycsdgecza4tkq3m5c7qs2r7m

GlaucoNet: Patch-Based Residual Deep Learning Network for Optic Disc and Cup Segmentation Towards Glaucoma Assessment

Rashmi Panda, N. B. Puhan, Bappaditya Mandal, Ganapati Panda
2021 SN Computer Science  
Damaged optic disc and optic cup assessment in color fundus image has been shown to be a promising method for glaucoma screening.  ...  The proposed patch-based deep network (GlaucoNet) is trained with a large number of preprocessed input image patches, big enough to include the important discriminatory information around each pixel.  ...  Sevastopolsky used a modification of U-Net convolutional neural network (CNN) for OD and OC segmentation in retinal images [37] .  ... 
doi:10.1007/s42979-021-00491-1 fatcat:eeaxifs5izdw7j7mmfgp3xpgsy

Dense Dilated Network with Probability Regularized Walk for Vessel Detection [article]

Lei Mou, Li Chen, Jun Cheng, Zaiwang Gu, Yitian Zhao, Jiang Liu
2019 arXiv   pre-print
To improve the connectivity of the segmented vessels, we also introduce a probability regularized walk algorithm to connect the broken vessels.  ...  The detection of retinal vessel is of great importance in the diagnosis and treatment of many ocular diseases. Many methods have been proposed for vessel detection.  ...  Many segmentation methods are based on the encoder-decoder structure, such as SegNet [23] and U-Net [22] .  ... 
arXiv:1910.12010v1 fatcat:zh7y42mwqzbrxmv2yt63ljeiaq

Automatic Retinal Vessel Segmentation via Deeply Supervised and Smoothly Regularized Network

Yi Lin, Honggong Zhang, Guang Hu
2018 IEEE Access  
In this paper, we propose an automatic retinal vessel segmentation network using deep supervision and smoothness regularization, which integrates holisticallynested edge detector (HED) and global smoothness  ...  INDEX TERMS Vessel segmentation, deep learning, medical image analysis, deep supervision, conditional random field.  ...  ACKNOWLEDGMENT The authors would like to thank the anonymous reviewers for their helpful suggestions.  ... 
doi:10.1109/access.2018.2844861 fatcat:nv5totuwevhxpli54suerajzye

Improving the performance of convolutional neural network for the segmentation of optic disc in fundus images using attention gates and conditional random fields

Bhargav J Bhatkalkar, Dheeraj R Reddy, Srikanth Prabhu, Sulatha V Bhandary
2020 IEEE Access  
In this paper, we are proposing a novel convolutional neural network architecture for the precise segmentation of the OD in fundus images.  ...  The localization and segmentation of optic disc (OD) in fundus images is a crucial step in the pipeline for detecting the early onset of retinal diseases, such as macular degeneration, diabetic retinopathy  ...  ACKNOWLEDGMENT We thank Kasturba Medical College (KMC), Manipal, India, for providing the fundus image dataset, and TensorFlow Research Cloud ( for providing the computing  ... 
doi:10.1109/access.2020.2972318 fatcat:b4k4pvupezgp5lw7zp7iwzaqzy

Deep Neural Architectures for Medical Image Semantic Segmentation: Review

Muhammad Zubair Khan, Mohan Kumar Gajendran, Yugyung Lee, Muazzam A. Khan
2021 IEEE Access  
The technique combines RetinaNet's one-stage detector with U-Net architecture for image segmentation.  ...  DRIVE Digital Retinal Images for Vessel Extraction (DRIVE) dataset is built up to empower comparative investigations on blood vessel segmentation in retinal fundus images.  ... 
doi:10.1109/access.2021.3086530 fatcat:hacpqwdxybh63j5ygebqszm7qq

Accurate Tumor Segmentation via Octave Convolution Neural Network

Bo Wang, Jingyi Yang, Jingyang Ai, Nana Luo, Lihua An, Haixia Feng, Bo Yang, Zheng You
2021 Frontiers in Medicine  
In this paper, we propose an effective and efficient method for tumor segmentation in liver CT images using encoder-decoder based octave convolution networks.  ...  Three-dimensional (3D) liver tumor segmentation from Computed Tomography (CT) images is a prerequisite for computer-aided diagnosis, treatment planning, and monitoring of liver cancer.  ...  Many medical image segmentation problems are improved based on U-Net.  ... 
doi:10.3389/fmed.2021.653913 pmid:34095168 pmcid:PMC8169966 fatcat:bqlgbbnhqvc3blskhgrzha62mm

Retinal Vessels Segmentation Based on Dilated Multi-Scale Convolutional Neural Network

Yun Jiang, Ning Tan, TingTing Peng, Hai Zhang
2019 IEEE Access  
Most methods based on deep convolutional neural network (DCNN) have small receptive fields, and hence they are unable to capture global context information of larger regions, with difficult to identify  ...  The final segmented retina vessels contain more noise with low classification accuracy. Therefore, in this paper, we propose a DCNN structure named as D-Net.  ...  In [27] , it used a Recurrent Convolutional Neural Network (RCNN) based on U-Net as well as a Recurrent Residual Convolutional Neural Network (RRCNN) based on U-Net models to segmentation retina vessel  ... 
doi:10.1109/access.2019.2922365 fatcat:462oythl2vb4benidwvrrtdo3m
« Previous Showing results 1 — 15 out of 290 results