Filters








8,459 Hits in 3.5 sec

Transfer Learning by Asymmetric Image Weighting for Segmentation across Scanners [article]

Veronika Cheplygina and Annegreet van Opbroek and M. Arfan Ikram and Meike W. Vernooij and Marleen de Bruijne
2017 arXiv   pre-print
Supervised learning has been very successful for automatic segmentation of images from a single scanner.  ...  The weight of each classifier is determined by the similarity between its training image and the test image.  ...  Acknowledgements This research was performed as part of the research project "Transfer learning in biomedical image analysis" which is financed by the Netherlands Organization for Scientific Research (  ... 
arXiv:1703.04981v1 fatcat:odbseupzq5hk3creftco4diioi

Asymmetric Ensemble of Asymmetric U-Net Models for Brain Tumor Segmentation With Uncertainty Estimation

Sarahi Rosas-Gonzalez, Taibou Birgui-Sekou, Moncef Hidane, Ilyess Zemmoura, Clovis Tauber
2021 Frontiers in Neurology  
In this study, an ensemble of two kinds of U-Net-like models based on both 3D and 2.5D convolutions is proposed to segment multimodal magnetic resonance images (MRI).  ...  Dice similarity coefficient for the whole tumor, tumor core, and tumor enhancing regions on BraTS 2019 validation dataset were 0.902, 0.815, and 0.773.  ...  Geraldine Brohan for the excellent secretary and administrative assistance.  ... 
doi:10.3389/fneur.2021.609646 pmid:34659077 pmcid:PMC8515181 fatcat:ryvi725mardchox5cwhilad2lq

3D Semi-Supervised Learning with Uncertainty-Aware Multi-View Co-Training [article]

Yingda Xia, Fengze Liu, Dong Yang, Jinzheng Cai, Lequan Yu, Zhuotun Zhu, Daguang Xu, Alan Yuille, Holger Roth
2020 arXiv   pre-print
We show the effectiveness of our proposed semi-supervised method on several public datasets from medical image segmentation tasks (NIH pancreas & LiTS liver tumor dataset).  ...  In addition, we propose an uncertainty-weighted label fusion mechanism to estimate the reliability of each view's prediction with Bayesian deep learning.  ...  Lingxi Xie, Siyuan Qiao and Yuyin Zhou for instructive discussions.  ... 
arXiv:1811.12506v2 fatcat:uf6zbwqo2bebri7d3z2qm3wmme

3D Semi-Supervised Learning with Uncertainty-Aware Multi-View Co-Training

Yingda Xia, Fengze Liu, Dong Yang, Jinzheng Cai, Lequan Yu, Zhuotun Zhu, Daguang Xu, Alan Yuille, Holger Roth
2020 2020 IEEE Winter Conference on Applications of Computer Vision (WACV)  
We show the effectiveness of our proposed semi-supervised method on several public datasets from medical image segmentation tasks (NIH pancreas & LiTS liver tumor dataset).  ...  In addition, we propose an uncertainty-weighted label fusion mechanism to estimate the reliability of each view's prediction with Bayesian deep learning.  ...  Lingxi Xie, Siyuan Qiao and Yuyin Zhou for instructive discussions.  ... 
doi:10.1109/wacv45572.2020.9093608 dblp:conf/wacv/XiaLYCYZXYR20 fatcat:4odj3l2nozcdpcbk2ela5ljp3q

Ensembling Low Precision Models for Binary Biomedical Image Segmentation [article]

Tianyu Ma, Hang Zhang, Hanley Ong, Amar Vora, Thanh D. Nguyen, Ajay Gupta, Yi Wang, Mert Sabuncu
2020 arXiv   pre-print
One of the major challenges for this task is that the appearance of foreground (positive) regions can be similar to background (negative) regions.  ...  Thus, in aggregate the false positive errors will cancel out, yielding high performance for the ensemble. Our strategy is general and can be applied with any segmentation model.  ...  For each scan, FLAIR, PD-weighted, T2-weighted, and T1-weighted images are provided.  ... 
arXiv:2010.08648v1 fatcat:vpo6zzc7l5ekxaah4hxchoorau

Orthogonal Ensemble Networks for Biomedical Image Segmentation [article]

Agostina J. Larrazabal, César Martínez, Jose Dolz, Enzo Ferrante
2021 arXiv   pre-print
The experimental results show that our approach produces more robust and well-calibrated ensemble models and can deal with challenging tasks in the context of biomedical image segmentation.  ...  We benchmark the proposed framework in two challenging brain lesion segmentation tasks --brain tumor and white matter hyper-intensity segmentation in MR images.  ...  Acknowledgments The authors gratefully acknowledge NVIDIA Corporation with the donation of the GPUs used for this research, and the support of UNL (CAID-0620190100145LI, CAID-50220140100084LI) and ANPCyT  ... 
arXiv:2105.10827v1 fatcat:tqu2pvifqjdfra4coaovxz5w2e

A Two-Stage Cascade Model with Variational Autoencoders and Attention Gates for MRI Brain Tumor Segmentation [article]

Chenggang Lyu, Hai Shu
2020 arXiv   pre-print
In this paper, we propose a two-stage encoder-decoder based model for brain tumor subregional segmentation.  ...  Automatic MRI brain tumor segmentation is of vital importance for the disease diagnosis, monitoring, and treatment planning.  ...  [12] , proposed a two-stage network, which used an asymmetrical U-Net, similar to Myronenko [14] , in the first stage to obtain a coarse prediction, and then employed a similar but wider network in  ... 
arXiv:2011.02881v2 fatcat:vthpqpa7ovcshc2xxfnl2adza4

3D MRI brain tumor segmentation using autoencoder regularization [article]

Andriy Myronenko
2018 arXiv   pre-print
Automated segmentation of brain tumors from 3D magnetic resonance images (MRIs) is necessary for the diagnosis, monitoring, and treatment planning of the disease.  ...  Here, we describe a semantic segmentation network for tumor subregion segmentation from 3D MRIs based on encoder-decoder architecture.  ...  Methods Our segmentation approach follows encoder-decoder based CNN architecture with an asymmetrically larger encoder to extract image features and a smaller decoder to reconstruct the segmentation mask  ... 
arXiv:1810.11654v3 fatcat:upbaid6myfekjddx4ako5m3cwq

Machine Learning Paradigm towards Content Based Image Retrieval on High Resolution Satellite Images

2019 VOLUME-8 ISSUE-10, AUGUST 2019, REGULAR ISSUE  
The learned SVM ensemble model is used to identify the images that most similar informative for active learning.  ...  In this paper, SVM based on random subspace and boosting ensemble learning is proposed for very high resolution satellite image retrieval.  ...  The learned SVM ensemble model is used to identify the images that most similar informative for active learning.  ... 
doi:10.35940/ijitee.b1104.1292s219 fatcat:phvtsg6kmnd45ebf6fisgwev6i

Uncertainty-aware multi-view co-training for semi-supervised medical image segmentation and domain adaptation

Yingda Xia, Dong Yang, Zhiding Yu, Fengze Liu, Jinzheng Cai, Lequan Yu, Zhuotun Zhu, Daguang Xu, Alan Yuille, Holger Roth
2020 Medical Image Analysis  
In this paper, we propose uncertainty-aware multi-view co-training (UMCT), a unified framework that addresses these two tasks for volumetric medical image segmentation.  ...  Experiments on the NIH pancreas segmentation dataset and a multi-organ segmentation dataset show state-of-the-art performance of the proposed framework on semi-supervised medical image segmentation.  ...  We propose to use asymmetric 3D models initialized with 2D pre-trained weights as the backbone network of each view to encourage diverse features for each view learning.  ... 
doi:10.1016/j.media.2020.101766 pmid:32623276 fatcat:qtygh6is5vci3fv6oqep4yqcn4

Redundancy Reduction in Semantic Segmentation of 3D Brain Tumor MRIs [article]

Md Mahfuzur Rahman Siddiquee, Andriy Myronenko
2021 arXiv   pre-print
for disease analysis and treatment planning.  ...  Given a set trained networks, we further introduce a confidence based ensembling techniques to further improve the performance.  ...  However, this alone does not often lead to a similar feature representation under perturbations, even if the final segmentation masks are similar.  ... 
arXiv:2111.00742v1 fatcat:cipz2s5igngrtcqg55j65qh2km

Multi-task Ensembles with Crowdsourced Features Improve Skin Lesion Diagnosis [article]

Ralf Raumanns, Elif K Contar, Gerard Schouten, Veronika Cheplygina
2020 arXiv   pre-print
The area under the receiver operating characteristic curve is 0.794 for the baseline model and 0.811 and 0.808 for multi-task ensembles respectively.  ...  ensemble strategies.  ...  Acknowledgments We would like to acknowledge all students who have contributed to this project with image annotation and/or code.  ... 
arXiv:2004.14745v2 fatcat:m4hubdtjdjafrom3gwrwm2ttj4

Ensemble Learning with Multiclassifiers on Pediatric Hand Radiograph Segmentation for Bone Age Assessment

Rui Liu, Yuanyuan Jia, Xiangqian He, Zhe Li, Jinhua Cai, Hao Li, Xiao Yang, Jyh-Cheng Chen
2020 International Journal of Biomedical Imaging  
We considered the hand segmentation task as a classification problem. The optimal segmentation threshold for each image was regarded as the prediction target.  ...  We utilized the normalized histogram, mean value, and variance of each image as input features to train the classification model, based on ensemble learning with multiple classifiers. 600 left-hand radiographs  ...  As for the impact of hand segmentation for BAA, it was obvious that there was a better performance in RMSE with the overall hand-segmented images, which suggested that the hand image segmentation step  ... 
doi:10.1155/2020/8866700 pmid:33178255 pmcid:PMC7609149 fatcat:acfgursmbrhuta4vh3rlrg2hve

Enhancing MRI Brain Tumor Segmentation with an Additional Classification Network [article]

Hieu T. Nguyen, Tung T. Le, Thang V. Nguyen, Nhan T. Nguyen
2020 arXiv   pre-print
Brain tumor segmentation plays an essential role in medical image analysis. In recent studies, deep convolution neural networks (DCNNs) are extremely powerful to tackle tumor segmentation tasks.  ...  On the BraTS's validation set, it achieved an average Dice score of 78.43%, 89.99%, and 84.22% respectively for the enhancing tumor, the whole tumor, and the tumor core.  ...  Acknowledgments This work was highly supported by Medical Imaging Department at Vingroup Big Data Institute (VinBigdata).  ... 
arXiv:2009.12111v2 fatcat:agxnfo4nvnd5pntju6up74tuta

Learning non-homogenous textures and the unlearning problem with application to drusen detection in retinal images

Noah Lee, Andrew F. Laine, Theodore R. Smith
2008 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro  
We perform probabilistic boosting and structural similarity clustering for fast selective learning in a large knowledge domain acquired over different time steps.  ...  In this work we present a novel approach for learning nonhomogenous textures without facing the unlearning problem.  ...  If we have a set of weak learners boosting builds a linear weighted ensemble hypothesis (3) For details on how to compute we refer to [1] .  ... 
doi:10.1109/isbi.2008.4541221 dblp:conf/isbi/LeeLS08 fatcat:f3cgrouolvht3hdexvk4d5skoy
« Previous Showing results 1 — 15 out of 8,459 results