Filters








1,653 Hits in 2.7 sec

Disentangling A Single MR Modality [article]

Lianrui Zuo, Yihao Liu, Yuan Xue, Shuo Han, Murat Bilgel, Susan M. Resnick, Jerry L. Prince, Aaron Carass
2022 arXiv   pre-print
Current methods learn disentangled representations using either paired multi-modal images with the same underlying anatomy or auxiliary labels (e.g., manual delineations) to provide inductive bias for  ...  Disentangling anatomical and contrast information from medical images has gained attention recently, demonstrating benefits for various image analysis tasks.  ...  Results show that our single-modal disentangling framework achieves performance comparable to methods which rely on multi-modal images for disentanglement.  ... 
arXiv:2205.04982v1 fatcat:srvmj4ad2vdpxcp2j7s5ymuy3e

Disentangled Representation Learning in Cardiac Image Analysis [article]

Agisilaos Chartsias, Thomas Joyce, Giorgos Papanastasiou, Michelle Williams, David Newby, Rohan Dharmakumar, Sotirios A. Tsaftaris
2019 arXiv   pre-print
We demonstrate that this high-level representation is ideally suited for several medical image analysis tasks, such as semi-supervised segmentation, multi-task segmentation and regression, and image-to-image  ...  Here, we explicitly learn this decomposed (disentangled) representation of imaging data, focusing in particular on cardiac images.  ...  Motivation Disentangled representations have considerable potential in the analysis of medical data.  ... 
arXiv:1903.09467v4 fatcat:lsdtpg2cove5thk4r35osw2gni

Disentangled representation learning in cardiac image analysis

Agisilaos Chartsias, Thomas Joyce, Giorgos Papanastasiou, Scott Semple, Michelle Williams, David E. Newby, Rohan Dharmakumar, Sotirios A. Tsaftaris
2019 Medical Image Analysis  
We demonstrate that this high-level representation is ideally suited for several medical image analysis tasks, such as semi-supervised segmentation, multi-task segmentation and regression, and image-to-image  ...  Here, we explicitly learn this decomposed (disentangled) representation of imaging data, focusing in particular on cardiac images.  ...  Motivation Disentangled representations have considerable potential in the analysis of medical data.  ... 
doi:10.1016/j.media.2019.101535 pmid:31351230 pmcid:PMC6815716 fatcat:amrltox6svgk7oemhjivkyzfly

Disentangle, align and fuse for multimodal and semi-supervised image segmentation [article]

Agisilaos Chartsias, Giorgos Papanastasiou, Chengjia Wang, Scott Semple, David E. Newby, Rohan Dharmakumar, Sotirios A. Tsaftaris
2020 arXiv   pre-print
Taking advantage of the common information shared between modalities (an organ's anatomy) is beneficial for multi-modality processing and learning.  ...  Despite advances in image analysis, we tend to treat each sequence, here termed modality, in isolation.  ...  However, a disentangled representation with modality invariant anatomy factors is not enough for multimodal learning.  ... 
arXiv:1911.04417v4 fatcat:qxlay6fzz5fdlcpta2epygydf4

Fine-Grained Action Retrieval Through Multiple Parts-of-Speech Embeddings [article]

Michael Wray, Diane Larlus, Gabriela Csurka, Dima Damen
2019 arXiv   pre-print
In this paper, we propose to enrich the embedding by disentangling parts-of-speech (PoS) in the accompanying captions. We build a separate multi-modal embedding space for each PoS tag.  ...  We also demonstrate the benefit of disentangling the PoS for the generic task of cross-modal video retrieval on the MSR-VTT dataset.  ...  We first disentangle a caption into its parts of speech (PoS) and learn a Multi-Modal Embedding Network (MMEN, Sec. 3.1) for each PoS (Sec. 3.2).  ... 
arXiv:1908.03477v1 fatcat:5po6jw4sczgzhezll4isf7x3my

Brain tumor segmentation with missing modalities via latent multi-source correlation representation [article]

Tongxue Zhou, Stéphane Canu, Pierre Vera, Su Ruan
2020 arXiv   pre-print
Multimodal MR images can provide complementary information for accurate brain tumor segmentation. However, it's common to have missing imaging modalities in clinical practice.  ...  Since there exists a strong correlation between multi modalities, a novel correlation representation block is proposed to specially discover the latent multi-source correlation.  ...  Modeling the latent multi-source correlation Inspired by a fact that, there is strong correlation between multi MR modalities, since the same scene (the same patient) is observed by different modalities  ... 
arXiv:2003.08870v4 fatcat:jlcwhpuzh5bjvnbes5xhyabvaa

Multi-task Shape Regression for Medical Image Segmentation [chapter]

Xiantong Zhen, Yilong Yin, Mousumi Bhaduri, Ilanit Ben Nachum, David Laidley, Shuo Li
2016 Lecture Notes in Computer Science  
different applications irrespective of modalities.  ...  In this paper, we propose a general segmentation framework of Multi-Task Shape Regression (MTSR) which formulates segmentation as multi-task learning to leverage its strength of jointly solving multiple  ...  Introduction Segmentation plays a fundamental role in medical image analysis, which however has long been regarded as a challenging task due to great diversity of applications in multiple modalities, huge  ... 
doi:10.1007/978-3-319-46726-9_25 fatcat:kha2d3adxrb57c6y3rzr6hx22q

Fine-Grained Action Retrieval Through Multiple Parts-of-Speech Embeddings

Michael Wray, Gabriela Csurka, Diane Larlus, Dima Damen
2019 2019 IEEE/CVF International Conference on Computer Vision (ICCV)  
In this paper, we propose to enrich the embedding by disentangling parts-of-speech (PoS) in the accompanying captions. We build a separate multi-modal embedding space for each PoS tag.  ...  We also demonstrate the benefit of disentangling the PoS for the generic task of cross-modal video retrieval on the MSR-VTT dataset.  ...  We first disentangle a caption into its parts of speech (PoS) and learn a Multi-Modal Embedding Network (MMEN, Sec. 3.1) for each PoS (Sec. 3.2).  ... 
doi:10.1109/iccv.2019.00054 dblp:conf/iccv/WrayCLD19 fatcat:7g5jonm73fegbhkuqrauidnz4a

Multi-Domain Image Completion for Random Missing Input Data [article]

Liyue Shen, Wentao Zhu, Xiaosong Wang, Lei Xing, John M. Pauly, Baris Turkbey, Stephanie Anne Harmon, Thomas Hogue Sanford, Sherif Mehralivand, Peter Choyke, Bradford Wood, Daguang Xu
2020 arXiv   pre-print
Specifically, we develop a novel multi-domain image completion method that utilizes a generative adversarial network (GAN) with a representational disentanglement scheme to extract shared skeleton encoding  ...  We further illustrate that the learned representation in multi-domain image completion could be leveraged for high-level tasks, e.g., segmentation, by introducing a unified framework consisting of image  ...  [39, 9, 3] also discuss how to extract representations from multi-modalities especially for segmentation with missing imaging modalities.  ... 
arXiv:2007.05534v1 fatcat:buih5jhb5javlla4mmz7v32eqm

Combining multimodal information for Metal Artefact Reduction: An unsupervised deep learning framework [article]

Marta B.M. Ranzini, Irme Groothuis, Kerstin Kläser, M. Jorge Cardoso, Johann Henckel, Sébastien Ourselin, Alister Hart, Marc Modat
2020 arXiv   pre-print
Given their different artefact appearance, their complementary information can compensate for the corrupted signal in either modality.  ...  We thus propose an unsupervised deep learning method for multimodal MAR. We introduce the use of Locally Normalised Cross Correlation as a loss term to encourage the fusion of multimodal information.  ...  First, the ADN architecture was not modified, keeping the same amount of parameters for either single-or multi-modality tests and thus not optimising the model capacity to the multimodal task.  ... 
arXiv:2004.09321v1 fatcat:llgyf3nllzd4bc4dvvehh5m3ri

Mind The Gap: Alleviating Local Imbalance for Unsupervised Cross-Modality Medical Image Segmentation [article]

Zixian Su, Kai Yao, Xi Yang, Qiufeng Wang, Yuyao Yan, Kaizhu Huang
2022 arXiv   pre-print
We conduct a series of experiments with two cross-modality adaptation tasks, i,e. cardiac substructure and abdominal multi-organ segmentation.  ...  Unsupervised cross-modality medical image adaptation aims to alleviate the severe domain gap between different imaging modalities without using the target domain label.  ...  Related Work Domain Adaptation in Medical Image Analysis. There has been a rapid development of deep learning models for medical image analysis [38, 42, 14, 15] .  ... 
arXiv:2205.11888v1 fatcat:a5z263udqbantdapyqnqmjvhse

Combining Multimodal Information for Metal Artefact Reduction: An Unsupervised Deep Learning Framework

Marta B.M. Ranzini, Irme Groothuis, Kerstin Klaser, M. Jorge Cardoso, Johann Henckel, Sebastien Ourselin, Alister Hart, Marc Modat
2020 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI)  
Given their different artefact appearance, their complementary information can compensate for the corrupted signal in either modality.  ...  We thus propose an unsupervised deep learning method for multimodal MAR. We introduce the use of Locally Normalised Cross Correlation as a loss term to encourage the fusion of multimodal information.  ...  First, the ADN architecture was not modified, keeping the same amount of parameters for either single-or multi-modality tests and thus not optimising the model capacity to the multimodal task.  ... 
doi:10.1109/isbi45749.2020.9098633 dblp:conf/isbi/RanziniGKCHOHM20 fatcat:fkw5hl2a2bcjha6gqcl4g6xqui

Review of Disentanglement Approaches for Medical Applications – Towards Solving the Gordian Knot of Generative Models in Healthcare [article]

Jana Fragemann, Lynton Ardizzone, Jan Egger, Jens Kleesiek
2022 arXiv   pre-print
Furthermore, we summarize the different notions of disentanglement, review approaches to disentangle latent space representations and metrics to evaluate the degree of disentanglement.  ...  Encouraging the latent representation of a generative model to be disentangled offers new perspectives of control and interpretability.  ...  Multi-Modal Brain Analysis Results from Synthetic MRI Modality Generation For Magnetic Resonance (MR) images, different contrast acquisitions show different aspects, important for diagnosis.  ... 
arXiv:2203.11132v1 fatcat:fxrniu6dtjcz5cumwientkqh7i

Information-based Disentangled Representation Learning for Unsupervised MR Harmonization [article]

Lianrui Zuo, Blake E. Dewey, Aaron Carass, Yihao Liu, Yufan He, Peter A. Calabresi, Jerry L. Prince
2021 arXiv   pre-print
CALAMITI learns a disentangled latent space using a unified structure for multi-site harmonization without the need for traveling subjects.  ...  In this work, we propose an unsupervised MR harmonization framework, CALAMITI (Contrast Anatomy Learning and Analysis for MR Intensity Translation and Integration), based on information bottleneck theory  ...  Introduction Magnetic resonance (MR) imaging is a commonly used non-invasive imaging modality due to its flexibility and good tissue contrast.  ... 
arXiv:2103.13283v1 fatcat:25l2fcymyba2pb4t3euokyvwdu

Deep Class-Specific Affinity-Guided Convolutional Network for Multimodal Unpaired Image Segmentation [article]

Jingkun Chen, Wenqi Li, Hongwei Li, Jianguo Zhang
2021 arXiv   pre-print
Multi-modal medical image segmentation plays an essential role in clinical diagnosis. It remains challenging as the input modalities are often not well-aligned spatially.  ...  In this paper, we propose an affinity-guided fully convolutional network for multimodal image segmentation.  ...  [12] used disentangled representations to achieve CT and MR adaptation.  ... 
arXiv:2101.01513v1 fatcat:po63pr66lzcbdlfu3w4fozdpjq
« Previous Showing results 1 — 15 out of 1,653 results