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
Radiology, Nuclear Medicine & Medical Imaging
Dingyi Lin et al.
Summary: Pancreatic fat accumulation is associated with various diseases, but its pathophysiology and imaging diagnostics have not received enough attention. This study used the nnU-Net model to automatically measure the distribution of pancreatic fat deposition on Dixon MRI in multicenter/population datasets. The results showed that the 3D dual-contrast model had the best performance, accurately assessing the distribution of pancreatic fat and demonstrating high reliability.
JOURNAL OF MAGNETIC RESONANCE IMAGING
(2023)
Article
Computer Science, Information Systems
Tao Lei et al.
Summary: Convolutional neural networks (CNNs) have achieved notable success in medical image segmentation, but suffer from a large number of parameters, making it difficult to deploy on low-power hardware. To address this issue, we propose a shape-guided ultralight network (SGU-Net) that reduces parameter count and improves segmentation accuracy using ultralight convolution and adversarial shape-constraint. Experimental results show that SGU-Net achieves higher segmentation accuracy with lower memory costs and outperforms state-of-the-art networks.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2023)
Article
Engineering, Biomedical
Yao Yao et al.
Summary: In this paper, a transferred DenseSE-Mask R-CNN (TDSMask R-CNN) Network segmentation model is proposed for pancreatic tumor segmentation. The model utilizes multi-scale features and attention mechanism to accurately obtain tumor regions in PET and MRI images, and alleviates network overfitting. Experimental results show that the proposed method achieves superior segmentation accuracy compared to existing methods.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Biology
Fei Wang et al.
Summary: In this study, a novel method for tumor segmentation of three-dimensional PET-CT images is proposed using Multi-modal Fusion and Calibration Networks (MFCNet). MFCNet effectively fuses the features of different modal images and calibrates the network on the tumor region, achieving better results compared to existing methods.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Computer Science, Interdisciplinary Applications
Georg Hille et al.
Summary: Deep learning-based segmentation of liver and hepatic lesions is increasingly important in clinical practice due to the rising incidence of liver cancer. This study proposes a hybrid network, SWTR-Unet, combining convolutional and Transformer architectures to accurately segment hepatic lesions in MRI. The results demonstrate that SWTR-Unet achieves comparable segmentation accuracy to manual expert segmentations in MRI and CT imaging based on correlation analysis and comparison with other networks.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2023)
Article
Computer Science, Interdisciplinary Applications
Xiahan Chen et al.
Summary: In this study, a model-driven deep learning method based on spiral transformation was proposed for the segmentation of pancreatic cancer. The method effectively applied 3D contextual information by mapping 3D images onto 2D planes while preserving the spatial relationship between textures. Additionally, a transformation-weight-corrected module and a smooth regularization based on rebuilding prior knowledge were embedded to optimize the segmentation results. Promising segmentation performance was achieved on multi-parametric MRIs according to extensive experiments.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2022)
Article
Chemistry, Analytical
Seok Oh et al.
Summary: The automatic segmentation of pancreatic cyst lesions on endoscopic ultrasonography (EUS) images was achieved using a deep-learning approach. The Attention U-Net model showed superior performance in dice similarity coefficient (DSC) and intersection over union (IoU) scores compared to other models in the internal test. However, there was no statistically significant difference between the Attention U-Net and the Basic U-Net model in the external test.
Article
Computer Science, Artificial Intelligence
Dingwen Zhang et al.
Summary: This study proposes a novel cross-modality deep feature learning framework for brain tumor segmentation from multi-modality MRI data. By incorporating cross-modality feature transition and fusion processes, the framework is able to effectively improve the performance of brain tumor segmentation.
PATTERN RECOGNITION
(2021)
Article
Oncology
Rebecca L. Siegel et al.
Summary: Every year, the American Cancer Society projects the numbers of new cancer cases and deaths in the United States, with the latest data showing a significant decline in lung cancer mortality, while prostate cancer mortality has plateaued and breast and colorectal cancer mortality have slowed. Improvements in treatment have accelerated progress against lung cancer, leading to a record drop in overall cancer mortality.
CA-A CANCER JOURNAL FOR CLINICIANS
(2021)
Article
Biochemical Research Methods
Fabian Isensee et al.
Summary: nnU-Net is a deep learning-based image segmentation method that automatically configures itself for diverse biological and medical image segmentation tasks, offering state-of-the-art performance as an out-of-the-box tool.
Review
Green & Sustainable Science & Technology
Xiangbin Liu et al.
Summary: Medical image segmentation based on deep learning has made significant contributions to sustainable medical care, but still faces challenges such as low segmentation accuracy and limited dataset size. Further research is needed to address these issues and improve the technology.
Article
Radiology, Nuclear Medicine & Medical Imaging
Sijuan Huang et al.
Summary: The study aims to create a network utilizing multi-sequence MRI for automatic contouring and compared its performance with manual human contouring. Results showed that the proposed network outperformed baseline models in all metrics. Additionally, it was found that three-sequence fusion (T1-T1DIXONC-T2) was superior to two-sequence fusion (T1-T2 and T1-T1DIXONC).
Article
Medicine, General & Internal
Lorraine Abel et al.
Summary: An algorithm based on a two-step nnU-Net architecture was developed for automated detection of pancreatic cystic lesions on CT scans, showing comparable performance to human readers. The algorithm had high sensitivity for large lesions and those located in the distal pancreas.
Article
Computer Science, Interdisciplinary Applications
Lingxi Xie et al.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2020)
Article
Medicine, General & Internal
Jonathan D Mizrahi et al.
Article
Radiology, Nuclear Medicine & Medical Imaging
Yuhua Chen et al.
Article
Computer Science, Artificial Intelligence
Yuri Sousa Aurelio et al.
NEURAL PROCESSING LETTERS
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Hykoush Asaturyan et al.
MACHINE LEARNING IN MEDICAL IMAGING (MLMI 2019)
(2019)
Article
Computer Science, Artificial Intelligence
Georgios Douzas et al.
EXPERT SYSTEMS WITH APPLICATIONS
(2018)
Article
Computer Science, Artificial Intelligence
Konstantinos Kamnitsas et al.
MEDICAL IMAGE ANALYSIS
(2017)
Article
Computer Science, Artificial Intelligence
Mohammad Havaei et al.
MEDICAL IMAGE ANALYSIS
(2017)
Article
Gastroenterology & Hepatology
Masao Tanaka et al.
Proceedings Paper
Optics
Konstantin Dmitriev et al.
MEDICAL IMAGING 2016: IMAGE PROCESSING
(2016)
Review
Gastroenterology & Hepatology
Brian K. P. Goh et al.
JOURNAL OF GASTROINTESTINAL SURGERY
(2014)
Article
Gastroenterology & Hepatology
Joshua A. Waters et al.
JOURNAL OF GASTROINTESTINAL SURGERY
(2008)