相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Computer Science, Artificial Intelligence
Xiaoqing Guo et al.
MEDICAL IMAGE ANALYSIS
(2022)
Article
Computer Science, Artificial Intelligence
Shunjie Dong et al.
Summary: In this paper, an enhanced Deformable U-Net (DeU-Net) is proposed for cardiac MRI segmentation. The DeU-Net consists of three modules: Temporal Deformable Aggregation Module (TDAM), Enhanced Deformable Attention Network (EDAN), and Probabilistic Noise Correction Module (PNCM). Experimental results demonstrate the state-of-the-art performance of DeU-Net on the Extended ACDC dataset.
MEDICAL IMAGE ANALYSIS
(2022)
Article
Computer Science, Artificial Intelligence
Vikas Verma et al.
Summary: Interpolation Consistency Training (ICT) is a simple and efficient algorithm for training Deep Neural Networks in the semi-supervised learning paradigm, which moves the decision boundary to low-density regions of the data distribution in classification problems. Experimental results show that ICT achieves state-of-the-art performance when applied to CIFAR-10 and SVHN benchmark datasets.
Article
Multidisciplinary Sciences
Chongshu Wu et al.
Summary: This study proposes a fully automatic pixel-wise semantic segmentation method based on pseudo-labels to reduce the manual labeling work and ensure the accuracy of segmentation. Experimental results show that the method performs well on the pathology image dataset and can be used to assist in clinical grading.
Article
Oncology
Chunling Zhang et al.
Summary: This article focuses on expanding the small data set in tumor segmentation based on deep learning. The proposed method includes image cutting and mirroring augmentation, segmentation of augmented images, and boundary reconstruction. Experimental results show that the proposed data augmentation method improves the accuracy and performance of tumor segmentation.
FRONTIERS IN ONCOLOGY
(2022)
Article
Computer Science, Interdisciplinary Applications
Tri Huynh et al.
Summary: In this study, a perturbation-based semi-supervised learning method called Adaptive Blended Consistency Loss (ABCL) is proposed to address the problem of class imbalance in medical image classification. The experiments demonstrate the effectiveness of ABCL in improving classification performance and outperforming other existing methods.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2022)
Article
Engineering, Biomedical
Hao Guan et al.
Summary: This paper surveys the recent advances of domain adaptation methods in medical image analysis, presents the motivation of introducing domain adaptation techniques to tackle domain heterogeneity issues in medical image analysis, and reviews the recent domain adaptation models in various medical image analysis tasks.
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Tariq Bdair et al.
Summary: Medical image segmentation is a major challenge in machine learning, and semi-supervised learning methods have been proposed to address the need for annotated data. This paper introduces ROAM, a semi-supervised learning method that generates virtual data through linear interpolation on randomly selected layers. By doing so, ROAM enhances the model's generalization ability and achieves state-of-the-art results in public datasets.
IET IMAGE PROCESSING
(2022)
Article
Computer Science, Artificial Intelligence
Yu Hua et al.
Summary: This paper proposes a semi-supervised method that narrows the gap between semi-supervised and fully supervised models by utilizing unlabeled data and establishing contrastive relationships between feature representation vectors through supervised contrastive learning. It overcomes data misuse and underutilization in semi-supervised frameworks, enhancing performance.
INTERNATIONAL JOURNAL OF NEURAL SYSTEMS
(2022)
Article
Computer Science, Interdisciplinary Applications
Saul Calderon-Ramirez et al.
Summary: The implementation of deep learning-based computer-aided diagnosis systems for classifying mammogram images can improve the accuracy, reliability, and cost of diagnosing patients. However, training deep learning models requires a large number of labeled images, which can be time-consuming and costly. To address this issue, publicly available datasets have been created for pre-training the models. However, using models trained on these datasets for transfer learning and fine-tuning with images from different hospitals or clinics may result in decreased performance due to mismatched distributions. In this study, a real-world scenario is evaluated using a novel target dataset sampled from a private clinic, with few labels and imbalanced data. The use of semi-supervised deep learning with transfer learning and data augmentation is shown to provide a meaningful advantage when dealing with scarce labeled observations.
MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING
(2022)
Article
Chemistry, Analytical
Khawla Brahim et al.
Summary: In this paper, the ICPIU-Net architecture is proposed for efficient segmentation of LV myocardium, myocardial infarction, and microvascular-obstructed tissues from LGE-MR images. The method incorporates inclusion and classification constraint networks to improve segmentation results and has shown significant agreement with manual contouring.
Article
Biochemical Research Methods
Zailiang Chen et al.
Summary: Optical coherence tomography angiography (OCTA) is an important noninvasive vascular imaging technique in vision-related diseases. However, the automatic segmentation of retinal vessels in OCTA is understudied and the existing methods require large-scale pixel-level annotated images, which is time-consuming. In this study, we propose a dual-consistency semi-supervised segmentation network incorporating multi-scale self-supervised puzzle subtasks (DCSS-Net) to tackle the challenge of limited annotations. Experimental results demonstrate the effectiveness of our approach.
BIOMEDICAL OPTICS EXPRESS
(2022)
Review
Health Care Sciences & Services
Gael Varoquaux et al.
Summary: Research in computer analysis of medical images has the potential to improve patients' health, but systematic challenges are impeding progress. From data limitations to research incentives, many issues hinder advancements. However, there are ongoing efforts to address these problems and recommendations for future improvement.
NPJ DIGITAL MEDICINE
(2022)
Article
Computer Science, Artificial Intelligence
Kaiping Wang et al.
Summary: Due to the difficulty in accessing a large amount of labeled data, semi-supervised learning is becoming an attractive solution in medical image segmentation. Apart from data-level and model-level consistency, this paper proposes utilizing auxiliary tasks and task-level consistency to excavate effective representations from unlabeled data for segmentation. Experimental results demonstrate the effectiveness of the proposed method.
MEDICAL IMAGE ANALYSIS
(2022)
Article
Computer Science, Artificial Intelligence
Xuxin Chen et al.
Summary: This paper reviews the recent studies on applying deep learning methods in medical image analysis, emphasizing the latest progress and contributions of state-of-the-art unsupervised and semi-supervised deep learning in this field. It also discusses major technical challenges and suggests possible solutions for future research efforts.
MEDICAL IMAGE ANALYSIS
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Linhu Liu et al.
Summary: BA-CHRLN is a novel network for semi-supervised 3D medical image segmentation, utilizing contrastive learning to share the encoder between two branches with supervised and unsupervised decoders. By introducing perturbation to enforce consistency and implementing a boundary-aware map, it enhances segmentation accuracy by capturing organ boundaries.
2022 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, ICIP
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Wenqiao Zhang et al.
Summary: In this paper, a novel semi-supervised learning framework called BoostMIS is proposed, which leverages adaptive pseudo labeling and informative active annotation to enhance the performance of medical image SSL models. By adaptively utilizing unlabeled data and finding informative samples, BoostMIS improves SSL label propagation and model training effectively.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Chenyu You et al.
Summary: This article introduces a novel method for contrastive voxel-wise representation learning in the context of medical image segmentation. By capturing 3D spatial context and rich anatomical information, this method effectively learns low-level and high-level features and inherits the benefits of hardness-aware property.
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION, MICCAI 2022, PT IV
(2022)
Article
Computer Science, Information Systems
Jialin Shi et al.
Summary: This paper investigates how to deal with the negative effects of noisy labels in medical image segmentation and proposes a hybrid noise-robust learning architecture. The method includes slice-level label-quality awareness, shape-awareness regularization loss, and a re-weighting strategy. Experiments show competitive performance of this hybrid architecture on two datasets.
Article
Computer Science, Artificial Intelligence
Shuihua Wang et al.
Summary: The advancement of biomedical imaging technologies leads to the daily generation of massive amounts of data, prompting the development of data fusion methods for better understanding. Apart from data generation, factors like noise, missing data, data scarcity, and high dimensionality also need to be considered in the fusion process.
INFORMATION FUSION
(2021)
Article
Computer Science, Artificial Intelligence
Yuhan Zhang et al.
Summary: The paper introduces a twin self-supervision based semi-supervised learning approach that combines two self-supervised strategies to learn from both labeled and unlabeled images. This method performs well in retinal anomaly classification based on SD-OCT images, achieving good classification performance with only a small percentage of labels.
Article
Computer Science, Artificial Intelligence
Hong-Yu Zhou et al.
Summary: This study introduces a novel Semi-Supervised Medical image Detector (SSMD), which provides effective supervision for unlabeled data through an adaptive consistency cost function and heterogeneous perturbation strategies to regularize predictions. Experimental results demonstrate that SSMD achieves state-of-the-art performance across various settings, with proposed modules showing strong capabilities through comprehensive ablation studies.
MEDICAL IMAGE ANALYSIS
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Xiangde Luo et al.
Summary: This paper introduces a novel framework for semi-supervised NPC GTV segmentation with Uncertainty Rectified Pyramid Consistency regularization. By supervising multi-scale pyramid predictions, the segmentation performance can be largely improved, achieving an average Dice score of 82.74% with 50% labeled images.
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2021, PT II
(2021)
Article
Computer Science, Artificial Intelligence
Georgios Kaissis et al.
Summary: PriMIA is a free, open-source software framework for privacy-preserving medical image analysis, performing well in instance testing and preventing data disclosure.
NATURE MACHINE INTELLIGENCE
(2021)
Article
Computer Science, Interdisciplinary Applications
Chenyu You et al.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2020)
Article
Computer Science, Interdisciplinary Applications
Davood Karimi et al.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2020)
Article
Biochemical Research Methods
Linfeng Yang et al.
PLOS COMPUTATIONAL BIOLOGY
(2020)
Proceedings Paper
Biochemical Research Methods
Jintai Chen et al.
2020 IEEE INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICINE
(2020)
Proceedings Paper
Engineering, Biomedical
Ruizhe Li et al.
2020 IEEE 17TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI 2020)
(2020)
Article
Computer Science, Artificial Intelligence
Takeru Miyato et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2019)
Article
Computer Science, Theory & Methods
Connor Shorten et al.
JOURNAL OF BIG DATA
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Gerda Bortsova et al.
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2019, PT VI
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Fernando Navarro et al.
MACHINE LEARNING IN MEDICAL IMAGING (MLMI 2019)
(2019)
Article
Computer Science, Interdisciplinary Applications
Olivier Bernard et al.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2018)
Article
Computer Science, Information Systems
Chenyu You et al.
Article
Computer Science, Interdisciplinary Applications
Catalina Tobon-Gomez et al.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2015)
