Filters








175 Hits in 9.4 sec

Weakly supervised segmentation of tumor lesions in PET-CT hybrid imaging

Marcel Früh, Marc Fischer, Andreas Schilling, Sergios Gatidis, Tobias Hepp
2021 Journal of Medical Imaging  
Purpose: We introduce and evaluate deep learning methods for weakly supervised segmentation of tumor lesions in whole-body fluorodeoxyglucose-positron emission tomography (FDG-PET) based solely on binary  ...  Results: A weakly supervised segmentation of tumor lesions was feasible with satisfactory performance [best median Dice score 0.47, interquartile range (IQR) 0.35] compared with a fully supervised U-Net  ...  We found that CAM, GradCAM++, and ScoreCAM are suitable CAM methods for weakly supervised segmentation as they capture the tumor lesions within PET images, and thus the inferior performance of weakly supervised  ... 
doi:10.1117/1.jmi.8.5.054003 pmid:34660843 pmcid:PMC8510879 fatcat:pvq5z4rhjbfe5ac6apremoi3de

Weakly Supervised PET Tumor Detection Using Class Response [article]

Amine Amyar, Romain Modzelewski, Pierre Vera, Vincent Morard, Su Ruan
2020 arXiv   pre-print
Using a weakly supervised learning is a promising way to address this problem, however, it is challenging to train one model to detect and locate efficiently different type of lesions due to the huge variation  ...  First, a simple convolutional neural network classifier is trained to predict the type of cancer on two 2D MIP images.  ...  Different studies tried to solve this problem by using a fully convolutional neural networks (FCNs) such as Network in Network (NN) [11] and GoogLeNet [15] .  ... 
arXiv:2003.08337v2 fatcat:ipm6x22nz5fabcnmnp4jay32ye

A Tetrahedron-Based Heat Flux Signature for Cortical Thickness Morphometry Analysis [chapter]

Yonghui Fan, Gang Wang, Natasha Lepore, Yalin Wang
2018 Lecture Notes in Computer Science  
in Multi-Parametric MRI 609 Deep nested level sets: Fully automated segmentation of cardiac MR images in patients with pulmonary hypertension 618 MS-Net: Mixed-Supervision Fully-Convolutional Networks  ...  for Emphysema Quantification 296 Semi-Automatic RECIST Labeling on CT Scans with Cascaded Convolutional Neural Networks 306 A Weakly-Supervised Learning-Based Feature Localization in Confocal Laser Endomicroscopy  ... 
doi:10.1007/978-3-030-00931-1_48 pmid:30338317 pmcid:PMC6191198 fatcat:dqhvpm5xzrdqhglrfftig3qejq

Weakly Supervised Tumor Detection in PET Using Class Response for Treatment Outcome Prediction

Amine Amyar, Romain Modzelewski, Pierre Vera, Vincent Morard, Su Ruan
2022 Journal of Imaging  
In this work, we propose a prediction framework including a 3D tumor segmentation in positron emission tomography (PET) images, based on a weakly supervised deep learning method, and an outcome prediction  ...  , respectively, and a dice coefficient of 73% and 0.77% for tumor segmentation.  ...  Therefore, we propose an approach based on a weakly supervised learning (WSL) , where image-level information is used to train a classifier based on a convolutional neural network (CNN) to predict the  ... 
doi:10.3390/jimaging8050130 fatcat:66oo6nou65bddc6yvfcynvtk7e

Front Matter: Volume 11313

Bennett A. Landman, Ivana Išgum
2020 Medical Imaging 2020: Image Processing  
The publisher is not responsible for the validity of the information or for any outcomes resulting from reliance thereon.  ...  Utilization of CIDs allows articles to be fully citable as soon as they are published online, and connects the same identifier to all online and print versions of the publication.  ...  information-embedded fully convolutional networks for multi-organ segmentation with improved data augmentation and instance normalization 11313 17Identification of kernels in a convolutional neural network  ... 
doi:10.1117/12.2570657 fatcat:be32besqknaybh6wibz7unuboa

Front Matter: Volume 12032

Ivana Išgum, Olivier Colliot
2022 Medical Imaging 2022: Image Processing  
Utilization of CIDs allows articles to be fully citable as soon as they are published online, and connects the same identifier to all online and print versions of the publication.  ...  segmentation from laparoscopic videos [12032-32] 0V Convolutional neural network with a hybrid loss function for fully automated segmentation of lymphoma lesions in FDG PET images [12032-33] 0W iv Proc  ...  convolutional long-short term memory networks [12032-115] 3B Weakly supervised brain tumor segmentation via semantic affinity deep neural network [12032-116] 3C Investigation of multi-cohort brain MRI  ... 
doi:10.1117/12.2638192 fatcat:ikfgnjefaba2tpiamxoftyi6sa

Toward High-Throughput Artificial Intelligence-Based Segmentation in Oncological PET Imaging [article]

Fereshteh Yousefirizi, Abhinav K. Jha, Julia Brosch-Lenz, Babak Saboury, Arman Rahmim
2021 arXiv   pre-print
Convolutional neural networks (CNNs) have shown impressive results and potential towards fully automated segmentation in medical imaging, and particularly PET imaging.  ...  for segmentation of tumors or normal organs in single and bi-modality scans.  ...  Ghassan Hamarneh and Kumar Abhishek from Simon Fraser University for very valuable discussions.  ... 
arXiv:2107.13661v4 fatcat:2cbb2vhaezfpxirbfl77i4pyjq

Deep Learning in Medical Ultrasound Image Segmentation: a Review [article]

Ziyang Wang
2021 arXiv   pre-print
In addition, common evaluation methods for image segmentation and ultrasound image segmentation datasets are summarized.  ...  In the end, the challenges and potential research directions for medical ultrasound image segmentation are discussed.  ...  Fully Convolutional Neural Networks for Segmentation Fully Convolutional Networks (FCN) is firstly introduced by [Long et al., 2015] , which is one of the most commonly used CNN-based neural networks  ... 
arXiv:2002.07703v3 fatcat:dosuiqzoh5e6tm4754wmxeifam

Medical Image Segmentation with 3D Convolutional Neural Networks: A Survey [article]

S Niyas, S J Pawan, M Anand Kumar, Jeny Rajan
2022 arXiv   pre-print
At present, convolutional neural networks (CNN) are the preferred choice for medical image analysis.  ...  Furthermore, the research gaps and future directions in 3D medical image segmentation are discussed.  ...  [100] proposed a weakly-supervised brain lesion segmentation using attentional representation learning from 3D image volumes with image-level annotation.  ... 
arXiv:2108.08467v3 fatcat:s2rzghycjbczpparmrflsdzujq

Application of Deep Learning in Automated Analysis of Molecular Images in Cancer: A Survey

Yong Xue, Shihui Chen, Jing Qin, Yong Liu, Bingsheng Huang, Hanwei Chen
2017 Contrast Media & Molecular Imaging  
Hence, in this paper, we review the applications of deep learning in molecular imaging in terms of tumor lesion segmentation, tumor classification, and survival prediction.  ...  We also outline some future directions in which researchers may develop more powerful deep learning models for better performance in the applications in cancer molecular imaging.  ...  [38] proposed a method based on weakly supervised stacked denoising autoencoders to segment brain lesion as well as reduce false positive.  ... 
doi:10.1155/2017/9512370 pmid:29114182 pmcid:PMC5661078 fatcat:ev3zrlx67vfo5mt23e5u3y2t64

Deep Semantic Segmentation of Natural and Medical Images: A Review [article]

Saeid Asgari Taghanaki, Kumar Abhishek, Joseph Paul Cohen, Julien Cohen-Adad, Ghassan Hamarneh
2020 arXiv   pre-print
sequenced models, weakly supervised, and multi-task methods and provide a comprehensive review of the contributions in each of these groups.  ...  Further, for each group, we analyze each variant of these groups and discuss the limitations of the current approaches and present potential future research directions for semantic image segmentation.  ...  Goceri (2019a) discussed the fully supervised, weakly supervised and transfer learning techniques for training deep neural networks for segmentation of medical images, and also discussed the existing  ... 
arXiv:1910.07655v3 fatcat:uxrrmb3jofcsvnkfkuhfwi62yq

A Survey of Cross-Modality Brain Image Synthesis [article]

Guoyang Xie, Jinbao Wang, Yawen Huang, Yefeng Zheng, Feng Zheng, Yaochu Jin
2022 arXiv   pre-print
Finally, we evaluate the challenges and provide several open directions for this community. All resources are available at https://github.com/M-3LAB/awesome-multimodal-brain-image-systhesis  ...  A realistic solution is to explore either an unsupervised learning or a semi-supervised learning to synthesize the absent neuroimaging data.  ...  Semi-Supervised Methods Guo et al. [2021] adopt a supervised method to train a lesion segmentation network.  ... 
arXiv:2202.06997v2 fatcat:kqxte2xrcrcpjfkkhwrcxdjqsu

AI-Based Detection, Classification and Prediction/Prognosis in Medical Imaging: Towards Radiophenomics [article]

Fereshteh Yousefirizi, Pierre Decazes, Amine Amyar, Su Ruan, Babak Saboury, Arman Rahmim
2022 arXiv   pre-print
This work reviews AI-based techniques, with a special focus on oncological PET and PET/CT imaging, for different detection, classification, and prediction/prognosis tasks.  ...  Radiomics analysis has the potential to be utilized as a noninvasive technique for the accurate characterization of tumors to improve diagnosis and treatment monitoring.  ...  [93] proposed a network composed of two parts for the detection of lymphomas and sFEPU. Their model consists of five fully connected convolutional neural networks (FCNs) in parallel.  ... 
arXiv:2110.10332v4 fatcat:vmpxhoolarbrve5ddyfn5umfim

Teacher-Student Architecture for Mixed Supervised Lung Tumor Segmentation [article]

Vemund Fredriksen, Svein Ole M. Svele, André Pedersen, Thomas Langø, Gabriel Kiss, Frank Lindseth
2021 arXiv   pre-print
Convolutional neural networks may be suited for such tasks, but require substantial amounts of labeled data to train. Obtaining labeled data is a challenge, especially in the medical domain.  ...  Purpose: Automating tasks such as lung tumor localization and segmentation in radiological images can free valuable time for radiologists and other clinical personnel.  ...  However, convolutional neural networks (CNNs) are memory intensive, especially for 3D volumes.  ... 
arXiv:2112.11541v1 fatcat:kuxnh7e3xndbfkj652nqazc3qy

Medical image analysis based on deep learning approach

Muralikrishna Puttagunta, S. Ravi
2021 Multimedia tools and applications  
It provides a systematic review of the articles for classification, detection, and segmentation of medical images based on DLA.  ...  Basicsof the principles and implementations of artificial neural networks and deep learning are essential for understanding medical image analysis in computer vision.  ...  Recurrent Neural Networks (RNNs) and convolutional neural networks are examples of supervised DL algorithms.  ... 
doi:10.1007/s11042-021-10707-4 pmid:33841033 pmcid:PMC8023554 fatcat:cm522go4nbdbnglgzpw4nu7tbi
« Previous Showing results 1 — 15 out of 175 results