62 Hits in 3.6 sec

UNETR: Transformers for 3D Medical Image Segmentation [article]

Ali Hatamizadeh, Yucheng Tang, Vishwesh Nath, Dong Yang, Andriy Myronenko, Bennett Landman, Holger Roth, Daguang Xu
2021 arXiv   pre-print
Inspired by the recent success of transformers for Natural Language Processing (NLP) in long-range sequence learning, we reformulate the task of volumetric (3D) medical image segmentation as a sequence-to-sequence  ...  Fully Convolutional Neural Networks (FCNNs) with contracting and expanding paths have shown prominence for the majority of medical image segmentation applications since the past decade.  ...  For 3D medical image segmentation, Xie et al.  ... 
arXiv:2103.10504v3 fatcat:wrxhzylwkzdaffdtal4dnckhkq

D-Former: A U-shaped Dilated Transformer for 3D Medical Image Segmentation [article]

Yixuan Wu, Kuanlun Liao, Jintai Chen, Jinhong Wang, Danny Z. Chen, Honghao Gao, Jian Wu
2022 arXiv   pre-print
Based on this design of Dilated Transformer, we construct a U-shaped encoder-decoder hierarchical architecture called D-Former for 3D medical image segmentation.  ...  Hence, recent work presented computer vision Transformer variants for medical image segmentation tasks and obtained promising performances.  ...  UNETR [49] proposed a 3D Transformercombining architecture for medical images, which treated Transformer layer as encoder to extract features and convolutional layer as decoder.  ... 
arXiv:2201.00462v2 fatcat:qewt25275bbzzongwb3jztqtwu

UNetFormer: A Unified Vision Transformer Model and Pre-Training Framework for 3D Medical Image Segmentation [article]

Ali Hatamizadeh, Ziyue Xu, Dong Yang, Wenqi Li, Holger Roth, Daguang Xu
2022 arXiv   pre-print
has led to state-of-the-art performance in various computer vision and medical image analysis tasks.  ...  In this work, we introduce a unified framework consisting of two architectures, dubbed UNetFormer, with a 3D Swin Transformer-based encoder and Convolutional Neural Network (CNN) and transformer-based  ...  Conclusion In this paper, we introduced a novel unified vision transformer-based framework for volumetric medical image segmentation.  ... 
arXiv:2204.00631v2 fatcat:4byyh7tln5egxbjlseojopfzxe

Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images [article]

Ali Hatamizadeh, Vishwesh Nath, Yucheng Tang, Dong Yang, Holger Roth, Daguang Xu
2022 arXiv   pre-print
In recent years, Fully Convolutional Neural Networks (FCNNs) approaches have become the de facto standard for 3D medical image segmentation.  ...  Inspired by the success of vision transformers and their variants, we propose a novel segmentation model termed Swin UNEt TRansformers (Swin UNETR).  ...  However, UNETR has shown better performance in terms of both accuracy and efficiency in different medical image segmentation tasks [16] .  ... 
arXiv:2201.01266v1 fatcat:kst2ddakorayfl4zfpimuacspy

SegTransVAE: Hybrid CNN – Transformer with Regularization for medical image segmentation [article]

Quan-Dung Pham
2022 arXiv   pre-print
Current research on deep learning for medical image segmentation exposes their limitations in learning either global semantic information or local contextual information.  ...  SegTransVAE is built upon encoder-decoder architecture, exploiting transformer with the variational autoencoder (VAE) branch to the network to reconstruct the input images jointly with segmentation.  ...  UNETR [4] leverages the power of transformers for volumetric medical image segmentation.  ... 
arXiv:2201.08582v3 fatcat:uz6lzazotvge7iag5xmwnajjfi

Self Pre-training with Masked Autoencoders for Medical Image Analysis [article]

Lei Zhou, Huidong Liu, Joseph Bae, Junjun He, Dimitris Samaras, Prateek Prasanna
2022 arXiv   pre-print
The segmentation and classification results reveal the promising potential of MAE self pre-training for medical image analysis.  ...  Masked Autoencoder (MAE) has recently been shown to be effective in pre-training Vision Transformers (ViT) for natural image analysis.  ...  UNETR is recently proposed for 3D image segmentation tasks such as brain tumor segmentation.  ... 
arXiv:2203.05573v1 fatcat:i24jbdx4jndz3kzdcgatpqqhmy

PHTrans: Parallelly Aggregating Global and Local Representations for Medical Image Segmentation [article]

Wentao Liu, Tong Tian, Weijin Xu, Huihua Yang, Xipeng Pan, Songlin Yan, Lemeng Wang
2022 arXiv   pre-print
Especially for medical image segmentation, many excellent hybrid architectures based on convolutional neural networks (CNNs) and Transformer have been presented and achieve impressive performance.  ...  In this paper, we propose a novel hybrid architecture for medical image segmentation called PHTrans, which parallelly hybridizes Transformer and CNN in main building blocks to produce hierarchical representations  ...  As a result, Transformer produces unsatisfactory performance in medical image segmentation.  ... 
arXiv:2203.04568v3 fatcat:3qaysebdtnaa5dpeqfccz42xym

BiTr-Unet: a CNN-Transformer Combined Network for MRI Brain Tumor Segmentation [article]

Qiran Jia, Hai Shu
2021 arXiv   pre-print
Convolutional neural networks (CNNs) have achieved remarkable success in automatically segmenting organs or lesions on 3D medical images.  ...  Therefore, we propose a CNN-Transformer combined model, called BiTr-Unet, with specific modifications for brain tumor segmentation on multi-modal MRI scans.  ...  Differences from UNETR A recently proposed network for 3D medical image segmentation, UNETR, also uses Transformer to extract long-range spatial dependencies [10] .  ... 
arXiv:2109.12271v2 fatcat:axnbiwkd5fd4vicfkdoh4hzgaq

High-Resolution Swin Transformer for Automatic Medical Image Segmentation [article]

Chen Wei, Shenghan Ren, Kaitai Guo, Haihong Hu, Jimin Liang
2022 arXiv   pre-print
The Resolution of feature maps is critical for medical image segmentation.  ...  Most of the existing Transformer-based networks for medical image segmentation are U-Net-like architecture that contains an encoder that utilizes a sequence of Transformer blocks to convert the input medical  ...  Medical Image Segmentation Since the excellent performance of U-Net architecture [28] for medical image segmentation, recently proposed Transformer-based medical image segmentation methods [21, 22,  ... 
arXiv:2207.11553v1 fatcat:5lg4z5r2xzhf7idybpk6rvblvu

Comparative Validation of AI and non-AI Methods in MRI Volumetry to Diagnose Parkinsonian Syndromes [article]

Joomee Song, Juyoung Hahm, Jisoo Lee, Chae Yeon Lim, Myung Jin Chung, Jinyoung Youn, Jin Whan Cho, Jong Hyeon Ahn, Kyung-Su Kim
2022 arXiv   pre-print
Automated segmentation and volumetry of brain magnetic resonance imaging (MRI) scans are essential for the diagnosis of Parkinson's disease (PD) and Parkinson's plus syndromes (P-plus).  ...  The segmentation times of V-Net and UNETR for the six brain structures per patient were 3.48 +- 0.17 and 48.14 +- 0.97 s, respectively, being at least 300 times faster than FS (15,735 +- 1.07 s).  ...  UNETR [27] is a transformer architecture for 3D medical-image segmentation.  ... 
arXiv:2207.11534v1 fatcat:3om5nolhvnhnzcapeujbktu6j4

Automatic Segmentation of Head and Neck Tumor: How Powerful Transformers Are? [article]

Ikboljon Sobirov, Otabek Nazarov, Hussain Alasmawi, Mohammad Yaqub
2022 arXiv   pre-print
This work shows that cancer segmentation via transformer-based models is a promising research area to further explore.  ...  Positron emission tomography and computed tomography are used to detect, segment and quantify the tumor region. Clinically, tumor segmentation is extensively time-consuming and prone to error.  ...  We also thank our colleagues Numan Saeed and Hashmat Shadab Malik for their fruitful discussions for the improvement of the paper.  ... 
arXiv:2201.06251v2 fatcat:mdp655gdfzgffgovzuzrwc6uwa

Hypernet-Ensemble Learning of Segmentation Probability for Medical Image Segmentation with Ambiguous Labels [article]

Sungmin Hong, Anna K. Bonkhoff, Andrew Hoopes, Martin Bretzner, Markus D. Schirmer, Anne-Katrin Giese, Adrian V. Dalca, Polina Golland, Natalia S. Rost
2021 arXiv   pre-print
Our approaches successfully estimated the segmentation probability maps that reflected the underlying structures and provided the intuitive control on segmentation for the challenging 3D medical image  ...  This challenge has been addressed by leveraging multiple annotations per image and the segmentation uncertainty.  ...  Configurations We compared the Residual UNet (Re-sUNet) [28, 53] and the multi-head transformer-based method (UNETR) [17] that showed the state-of-the-art performance for 3D medical images to our methods  ... 
arXiv:2112.06693v1 fatcat:hopo6jzvhndhbnnl5lduivvy6i

Deep Segmentation Networks for Segmenting Kidneys and Detecting Kidney Stones in Unenhanced Abdominal CT Images

Dan Li, Chuda Xiao, Yang Liu, Zhuo Chen, Haseeb Hassan, Liyilei Su, Jun Liu, Haoyu Li, Weiguo Xie, Wen Zhong, Bingding Huang
2022 Diagnostics  
Recent breakthroughs of deep learning algorithms in medical imaging, automated detection, and segmentation techniques for renal (kidney) in abdominal computed tomography (CT) images have been limited.  ...  Upon comparison, the models trained with the proposed training scheme enable the highly accurate 2D and 3D segmentation of kidneys and kidney stones.  ...  University for reviewing the annotated dataset.  ... 
doi:10.3390/diagnostics12081788 pmid:35892498 pmcid:PMC9330428 fatcat:amkihaminfef3mu4b6rpgwx7oe

3D Shuffle-Mixer: An Efficient Context-Aware Vision Learner of Transformer-MLP Paradigm for Dense Prediction in Medical Volume [article]

Jianye Pang, Cheng Jiang, Yihao Chen, Jianbo Chang, Ming Feng, Renzhi Wang, Jianhua Yao
2022 arXiv   pre-print
In this paper, we propose a novel 3D Shuffle-Mixer network of a new Local Vision Transformer-MLP paradigm for medical dense prediction.  ...  Therefore, designing an elegant and efficient vision transformer learner for dense prediction in medical volume is promising and challenging.  ...  CNN-Transformer Backbones in 3D Medical Image Analysis: 3D segmentation is one of the most general and representative tasks in medical image analysis.  ... 
arXiv:2204.06779v1 fatcat:txccldfcdfesjetd4f7vxzskqq

TransConver: Transformer and convolution parallel network for developing automatic brain tumor segmentation in MRI images

Junjie Liang, Cihui Yang, Mengjie Zeng, Xixi Wang
2021 Quantitative Imaging in Medicine and Surgery  
Medical image segmentation plays a vital role in computer-aided diagnosis (CAD) systems.  ...  In this paper, we proposed TransConver, a U-shaped segmentation network based on convolution and transformer for automatic and accurate brain tumor segmentation in MRI images.  ...  Therefore, how to combine CNN and transformer is an essential challenge for medical image segmentation.  ... 
doi:10.21037/qims-21-919 pmid:35371952 pmcid:PMC8923874 fatcat:p33tfjvlinfcrjqptvvvh6fqce
« Previous Showing results 1 — 15 out of 62 results