Filters








44 Hits in 6.0 sec

Source-Free Domain Adaptive Fundus Image Segmentation with Denoised Pseudo-Labeling [article]

Cheng Chen, Quande Liu, Yueming Jin, Qi Dou, Pheng-Ann Heng
2021 arXiv   pre-print
Experimental results on cross-domain fundus image segmentation show that without using any source images or altering source training, our approach achieves comparable or even higher performance than state-of-the-art  ...  We present a novel denoised pseudo-labeling method for this problem, which effectively makes use of the source model and unlabeled target data to promote model self-adaptation from pseudo labels.  ...  In this paper, we present a novel denoised pseudo-labeling (DPL) method for source-free unsupervised domain adaptation.  ... 
arXiv:2109.09735v1 fatcat:xzvlngqdbjbx7cpwxdddyivzh4

Target and Task specific Source-Free Domain Adaptive Image Segmentation [article]

Vibashan VS, Jeya Maria Jose Valanarasu, Vishal M. Patel
2022 arXiv   pre-print
Therefore, adapting the source model with noisy pseudo labels reduces its segmentation capability while addressing the domain shift.  ...  To this end, we propose a two-stage approach for source-free domain adaptive image segmentation: 1) Target-specific adaptation followed by 2) Task-specific adaptation.  ...  Reducing the entropy or denoising pseudo labels helps close the domain gap between the source and target data.  ... 
arXiv:2203.15792v1 fatcat:vwlupjizyjhp7pbnf7ofz3trnm

Cross-denoising Network against Corrupted Labels in Medical Image Segmentation with Domain Shift

Qinming Zhang, Luyan Liu, Kai Ma, Cheng Zhuo, Yefeng Zheng
2020 Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  
In this paper, we propose a novel robust cross-denoising framework using two peer networks to address domain shift and corrupted label problems with a peer-review strategy.  ...  Deep convolutional neural networks (DCNNs) have contributed many breakthroughs in segmentation tasks, especially in the field of medical imaging.  ...  (OC) segmentation tasks with domain shift and noisy labels. proposed a plug-and-play domain adaptation module (DAM) by adapting the source and target domains in the feature space, to solve the cardiac  ... 
doi:10.24963/ijcai.2020/146 dblp:conf/ijcai/ZhangLMZZ20 fatcat:vtl34xuwojcy5l6pmlytxpnq6q

Cross-denoising Network against Corrupted Labels in Medical Image Segmentation with Domain Shift [article]

Qinming Zhang, Luyan Liu, Kai Ma, Cheng Zhuo, Yefeng Zheng
2020 arXiv   pre-print
In this paper, we propose a novel robust cross-denoising framework using two peer networks to address domain shift and corrupted label problems with a peer-review strategy.  ...  Deep convolutional neural networks (DCNNs) have contributed many breakthroughs in segmentation tasks, especially in the field of medical imaging.  ...  For medical image segmentation, Dou et al. [2018] proposed a plug-and-play domain adaptation module (DAM) by adapting the source and target domains in the feature space, to solve the cardiac structure  ... 
arXiv:2006.10990v1 fatcat:4pfw3n4wu5he7krp57yp3lrtry

Source-free unsupervised domain adaptation for cross-modality abdominal multi-organ segmentation [article]

Jin Hong, Yu-Dong Zhang, Weitian Chen
2022 arXiv   pre-print
The pseudo-labels output from the top segmentation network are used to guide the style compensation network to generate source-like images.  ...  Therefore, we propose an effective source-free unsupervised domain adaptation method for cross-modality abdominal multi-organ segmentation without source dataset access.  ...  In the medical image analysis field, Chen et al. (2021) proposed a denoised pseudo-labeling strategy for source-free unsupervised domain adaptation in fundus image segmentation.  ... 
arXiv:2111.12221v4 fatcat:medqgshumfdijbesm6kzzyyche

On-the-Fly Test-time Adaptation for Medical Image Segmentation [article]

Jeya Maria Jose Valanarasu, Pengfei Guo, Vibashan VS, Vishal M. Patel
2022 arXiv   pre-print
During test-time, the model takes in just the new test image and generates a domain code to adapt the features of source model according to the test data.  ...  To achieve this, we propose a new framework called Adaptive UNet where each convolutional block is equipped with an adaptive batch normalization layer to adapt the features with respect to a domain code  ...  In [6] , a label-free entropy loss is defined over target distribution with a domain-invariant prior. In [7] , an uncertainty aware denoised pseudo label method is proposed.  ... 
arXiv:2203.05574v1 fatcat:vrsu2cldmnb75hielfilpxpr2e

Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation [article]

Nima Tajbakhsh, Laura Jeyaseelan, Qian Li, Jeffrey Chiang, Zhihao Wu, Xiaowei Ding
2020 arXiv   pre-print
Recently, a large body of research has studied the problem of medical image segmentation with imperfect datasets, tackling two major dataset limitations: scarce annotations where only limited annotated  ...  The medical imaging literature has witnessed remarkable progress in high-performing segmentation models based on convolutional neural networks.  ...  Domain Adaptation with Target Labels: If the segmentation masks are available for both domains, there is no longer a distinction between the choice of source and target domains.  ... 
arXiv:1908.10454v2 fatcat:mjvfbhx75bdkbheysq3r7wmhdi

Appearance invariance in convolutional networks with neighborhood similarity [article]

Tolga Tasdizen, Mehdi Sajjadi, Mehran Javanmardi, Nisha Ramesh
2017 arXiv   pre-print
We also demonstrate its advantages for digit recognition, semantic labeling and cell detection problems.  ...  We present a neighborhood similarity layer (NSL) which induces appearance invariance in a network when used in conjunction with convolutional layers.  ...  Co-training [15] which uses classifiers with different views to generate pseudo-labels has been used for domain adaptation [3] .  ... 
arXiv:1707.00755v1 fatcat:qtplz5tnozd6nh2rzdlr5bfpka

State-of-the-art in retinal optical coherence tomography image analysis

Ahmadreza Baghaie, Zeyun Yu, Roshan M D'Souza
2015 Quantitative Imaging in Medicine and Surgery  
Another major step in OCT image analysis involves the use of segmentation techniques for distinguishing between different structures, especially in retinal OCT volumes.  ...  image registration techniques.  ...  This threshold can be a constant or even better, sub-band adaptive or spatially adaptive.  ... 
doi:10.3978/j.issn.2223-4292.2015.07.02 pmid:26435924 pmcid:PMC4559975 fatcat:m35jwvfrlfamrntdfmqqqwp5oi

Self-supervised learning methods and applications in medical imaging analysis: A survey [article]

Saeed Shurrab, Rehab Duwairi
2021 arXiv   pre-print
The availability of high quality annotated medical imaging datasets is a major problem that collides with machine learning applications in the field of medical imaging analysis and impedes its advancement  ...  This article reviews the state-of-the-art research directions in self-supervised learning approaches for image data with concentration on their applications in the field of medical imaging analysis.  ...  with denoised images.  ... 
arXiv:2109.08685v2 fatcat:iu2zanqqrnaflawcxndb6xszgu

Vision Transformers in Medical Computer Vision – A Contemplative Retrospection [article]

Arshi Parvaiz, Muhammad Anwaar Khalid, Rukhsana Zafar, Huma Ameer, Muhammad Ali, Muhammad Moazam Fraz
2022 arXiv   pre-print
Along with this, we also demystify several imaging modalities used in Medical Computer Vision.  ...  Recent escalation in the field of computer vision underpins a huddle of algorithms with the magnificent potential to unravel the information contained within images.  ...  [234] proposed a spatial adaptive and transformer fusion network (STFNet) for denoising low count PET with MRI. They adapted dual path using the spatial-adaptive block to extract features.  ... 
arXiv:2203.15269v1 fatcat:wecjpoikbvfz5cygytqpktoxdq

Table of Contents

2021 IEEE Signal Processing Letters  
Wang Multi-Label Classification of Fundus Images With Graph Convolutional Network and Self-Supervised Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ...  Zhao Study on Convergence of Plug-and-Play ISTA With Adaptive-Kernel Denoisers . . . . . . . . . . . . T. Liu, L. Xing, and Z.  ... 
doi:10.1109/lsp.2021.3134549 fatcat:m6obtl7k7zdqvd62eo3c4tptfy

2021 Index IEEE Signal Processing Letters Vol. 28

2021 IEEE Signal Processing Letters  
., +, LSP 2021 66-70 Multi-Label Classification of Fundus Images With Graph Convolutional Network and Self-Supervised Learning.  ...  Parashar, D., +, LSP 2021 66-70 Multi-Label Classification of Fundus Images With Graph Convolutional Network and Self-Supervised Learning.  ... 
doi:10.1109/lsp.2022.3145253 fatcat:a3xqvok75vgepcckwnhh2mty74

Convex Shape Prior for Deep Neural Convolution Network based Eye Fundus Images Segmentation [article]

Jun Liu, Xue-Cheng Tai, Shousheng Luo
2020 arXiv   pre-print
Convex Shapes (CS) are common priors for optic disc and cup segmentation in eye fundus images. It is important to design proper techniques to represent convex shapes.  ...  As an application example, we apply the convexity prior layer to the retinal fundus images segmentation by taking the popular DeepLabV3+ as a backbone network.  ...  Besides, our method does not use any domain adaptation technique, so one can integrate our method with the domain adaptation method to further improve the generalization ability for retinal images segmentation  ... 
arXiv:2005.07476v1 fatcat:sxxdooobefhr5ikvpn6yf3e3tu

An overview of deep learning in medical imaging focusing on MRI

Alexander Selvikvåg Lundervold, Arvid Lundervold
2018 Zeitschrift für Medizinische Physik  
imaging by pointing out good educational resources, state-of-the-art open-source code, and interesting sources of data and problems related medical imaging.  ...  to image retrieval, from segmentation to disease prediction; (iii) provide a starting point for people interested in experimenting and perhaps contributing to the field of machine learning for medical  ...  Our work was financially supported by the Bergen Research Foundation through the project "Computational medical imaging and machine learning -methods, infrastructure and applications".  ... 
doi:10.1016/j.zemedi.2018.11.002 fatcat:kkimovnwcrhmth7mg6h6cpomjm
« Previous Showing results 1 — 15 out of 44 results