检索结果-内蒙古大学图书馆

21st Pacific Rim International Conference on Artificial Intelligence, PRICAI 2024

作者： Shao, Liangjing Chen, Benshuang Chen, Xinrong Academy for Engineering and Technology Fudan University Shanghai200438 China Shanghai Key Laboratory for Medical Imaging Computing and Computer Assisted Intervention Fudan University Shanghai200032 China

ISBN: (纸本)9789819601271

Vision-based real-time pose tracking of endoscope is significant for navigation and automation of endoscopy. In this work, a deep learning-based framework is proposed for vision-based real-time pose tracking of nasal endoscope based on endoscopic images. Specifically, an attention-based module is introduced into the feature encoder to extract and integrate multi-dimensional and multi-scale information. Furthermore, a novel pose encoder is proposed to extract more complicated representation from the feature map for pose vector prediction. The proposed method is compared with state-of-the-art methods on the dataset, which consists of endoscopic videos and endoscope poses obtaining by an optical tracking system. The experimental results demonstrate that the proposed network achieves better performance than state-of-the-art works for real-time nasal endoscope pose estimation. The average tracking error of our method is 5.78 mm, while the relative translation error and rotation error are respectively 0.48 mm and 0.22 degree on average. In addition, the average inference speed of our network is 46 fps, which achieves the real-time requirement. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Endoscopy

来源：评论

学校读者我要写书评

暂无评论

REMOTE: Real-time Ego-motion Tracking for Various Endoscopes via Multimodal Visual Feature Learning

arXiv

引用

arXiv 2025年

作者： Shao, Liangjing Chen, Benshuang Zhao, Shuting Chen, Xinrong Academy for Engineering & Technology Fudan University China Shanghai Key Laboratory of Medical Image Computing and Computer-Assisted Intervention Fudan University China

Real-time ego-motion tracking for endoscope is a significant task for efficient navigation and robotic automation of endoscopy. In this paper, a novel framework is proposed to perform real-time ego-motion tracking for endoscope. Firstly, a multi-modal visual feature learning network is proposed to perform relative pose prediction, in which the motion feature from the optical flow, the scene features and the joint feature from two adjacent observations are all extracted for prediction. Due to more correlation information in the channel dimension of the concatenated image, a novel feature extractor is designed based on an attention mechanism to integrate multi-dimensional information from the concatenation of two continuous frames. To extract more complete feature representation from the fused features, a novel pose decoder is proposed to predict the pose transformation from the concatenated feature map at the end of the framework. At last, the absolute pose of endoscope is calculated based on relative poses. The experiment is conducted on three datasets of various endoscopic scenes and the results demonstrate that the proposed method outperforms state-of-the-art methods. Besides, the inference speed of the proposed method is over 30 frames per second, which meets the real-time requirement. The project page is here: *** © 2025, CC BY.

关键词： Endoscopy

来源：评论

学校读者我要写书评

暂无评论

NaMA-Mamba: Foundation model for generalizable nasal disease detection using masked autoencoder with Mamba on endoscopic images

引用

computerized medical Imaging and Graphics 2025年 122卷 102524-102524页

作者： Wang, Wensheng Jin, Zewen Liu, Xueli Chen, Xinrong Academy for Engineering and Technology Fudan University Shanghai200433 China Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention Shanghai200032 China Eye & ENT Hospital of Fudan University Shanghai200031 China

Artificial intelligence (AI) has shown great promise in analyzing nasal endoscopic images for disease detection. However, current AI systems require extensive expert-labeled data for each specific medical condition, limiting their applications. In this work, the challenge is addressed through two key innovations, the creation of the first large-scale pre-training dataset of nasal endoscopic images, and the development of a novel self-learning AI system specifically designed for nasal endoscopy, named NaMA-Mamba. In the proposed NaMA-Mamba model, two key technologies are utilized, which are the nasal endoscopic state space model (NE-SSM) for analyzing sequences of images and an enhanced learning mechanism (CoMAE) for capturing fine details in nasal tissues. These innovations enable the system to learn effectively from unlabeled images while maintaining high accuracy across different diagnostic tasks. In extensive testing, NaMA-Mamba achieved remarkable results using minimal labeled data, matching the performance of traditional systems that require full expert labeling while needing only 1% of the labeled data for tasks such as detecting nasal polyps and identifying nasopharyngeal cancer. These results demonstrate the potential of NaMA-Mamba to significantly improve the efficiency and accessibility of AI-assisted nasal disease diagnosis in clinical practice. © 2025 Elsevier Ltd

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

DM-Mamba: Dual-domain Multi-scale Mamba for MRI Reconstruction

arXiv

引用

arXiv 2025年

作者： Meng, Yucong Yang, Zhiwei Song, Zhijian Shi, Yonghong Fu, Kexue Digital Medical Research Center School of Basic Medical Science Fudan University Shanghai200032 China Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention Shanghai200032 China Academy of Engineering and Technology Fudan University Shanghai200433 China Jinan China

The accelerated MRI reconstruction poses a challenging ill-posed inverse problem due to the significant undersampling in k-space. Deep neural networks, such as CNNs and ViT, have shown substantial performance improvements for this task while encountering the dilemma between global receptive fields and efficient computation. To this end, this paper pioneers exploring Mamba, a new paradigm for long-range dependency modeling with linear complexity, for efficient and effective MRI reconstruction. However, directly applying Mamba to MRI reconstruction faces three significant issues: (1) Mamba’s row-wise and column-wise scanning disrupts k-space’s unique spectrum, leaving its potential in k-space learning unexplored. (2) Existing Mamba methods unfold feature maps with multiple lengthy scanning paths, leading to long-range forgetting and high computational burden. (3) Mamba struggles with spatially-varying contents, resulting in limited diversity of local representations. To address these, we propose a dual-domain multi-scale Mamba for MRI reconstruction from the following perspectives: (1) We pioneer vision Mamba in k-space learning. A circular scanning is customized for spectrum unfolding, benefiting the global modeling of k-space. (2) We propose a multi-scale Mamba with an efficient scanning strategy in both image and k-space domains. It mitigates long-range forgetting and achieves a better trade-off between efficiency and performance. (3) We develop a local diversity enhancement module to improve the spatially-varying representation of Mamba. Extensive experiments are conducted on three public datasets for MRI reconstruction under various undersampling patterns. Comprehensive results demonstrate that our method significantly outperforms state-of-the-art methods with lower computational cost. Implementation code will be available in https://***/XiaoMengLiLiLi/DM-Mamba. Copyright © 2025, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Exploring CLIP’s Dense Knowledge for Weakly Supervised Semantic Segmentation

arXiv

引用

arXiv 2025年

作者： Yang, Zhiwei Meng, Yucong Fu, Kexue Tang, Feilong Wang, Shuo Song, Zhijian Academy for Engineering and Technology Fudan University Shanghai200433 China Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention China Digital Medical Research Center School of Basic Medical Sciences Fudan University China Shandong Computer Science Center National Supercomputer Center in Jinan China

Weakly Supervised Semantic Segmentation (WSSS) with image-level labels aims to achieve pixel-level predictions using Class Activation Maps (CAMs). Recently, Contrastive Language-image Pre-training (CLIP) has been introduced in WSSS. However, recent methods primarily focus on image-text alignment for CAM generation, while CLIP’s potential in patch-text alignment remains unexplored. In this work, we propose ExCEL to explore CLIP’s dense knowledge via a novel patch-text alignment paradigm for WSSS. Specifically, we propose Text Semantic Enrichment (TSE) and Visual Calibration (VC) modules to improve the dense alignment across both text and vision modalities. To make text embeddings semantically informative, our TSE module applies Large Language Models (LLMs) to build a dataset-wide knowledge base and enriches the text representations with an implicit attribute-hunting process. To mine fine-grained knowledge from visual features, our VC module first proposes Static Visual Calibration (SVC) to propagate fine-grained knowledge in a non-parametric manner. Then Learnable Visual Calibration (LVC) is further proposed to dynamically shift the frozen features towards distributions with diverse semantics. With these enhancements, ExCEL not only retains CLIP’s training-free advantages but also significantly outperforms other state-of-the-art methods with much less training cost on PASCAL VOC and MS COCO. Code is available at https://***/zwyang6/ExCEL. Copyright © 2025, The Authors. All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

A Contrast-Enhanced Null Subtraction Imaging Method using Dynamic DC Bias in Ultrafast Ultrasound imaging

引用

IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control 2025年 PP卷 PP页

作者： Yan, Xin Yang, Xiaodong Jing, Lingling Guo, Wei Wang, Yingqi Su, Xinwei Wang, Yuanyuan Shenyang University of Technology School of Information Science and Engineering Shenyang110870 China Ultrasound Department of Anshan Central Hospital Anshan114000 China CO. Ltd Suzhou215123 China Fudan University Department of Electronic Engineering and the Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention of Shanghai Shanghai200433 China

Enhancing the resolution and contrast of ultrafast ultrasound imaging is imperative for the accuracy of clinical diagnostics. Null subtraction imaging (NSI) is a nonlinear beamforming technique capable of significantly enhancing lateral resolution. However, it suffers from issues of low-quality speckle pattern and poor contrast performance. To address this issue, we propose a novel contrast-enhanced NSI method that utilizes dynamic DC bias. Innovatively, we construct the dynamic DC bias using a generalized coherence factor (GCF) and a sigmoid transformation function that adapts the DC value based on the signal characteristics of different imaging regions. Furthermore, a normalization scheme is proposed to optimize the beamforming output, ensuring uniform pixel intensity throughout the final image. Simulation, phantom, and in vivo data are utilized for ultrasound beamforming to evaluate the performance of the proposed method. Quantitative results show that the proposed method significantly enhances the contrast ratio (CR) by 197%, the contrast-to-noise ratio (CNR) by 341%, the speckle signal-to-noise ratio (sSNR) by 302%, and the generalized contrast-to-noise ratio (gCNR) by 106% compared to the original NSI (in phantom). Point target imaging results indicate that the proposed method achieves a main lobe width slightly wider than the original NSI method, but much narrower than those of DAS and GCF. These findings confirm that the proposed method significantly enhances imaging contrast while preserving high resolution, which is of great significance for the further clinical application of ultrafast ultrasound imaging. © 1986-2012 IEEE.

关键词： Phantoms

来源：评论

学校读者我要写书评

暂无评论

Anatomy-inspired model for critical landmark localization in 3D spinal ultrasound volume data

引用

medical image Analysis 2025年 103卷

作者： Huang, Yi Jiao, Jing Yu, Jinhua Zheng, Yongping Wang, Yuanyuan Biomedical Engineering Center Fudan University Shanghai 200433 China Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention of Shanghai Fudan University 200433 China Department of Biomedical Engineering The Hong Kong Polytechnic University SAR Hong Kong China Research Institute for Smart Ageing The Hong Kong Polytechnic University SAR Hong Kong China

Three-dimensional (3D) spinal ultrasound imaging has demonstrated its promising potential in measuring spinal deformity through recent studies, and it is more suitable for massive early screening and longitudinal follow-up of adolescent idiopathic scoliosis (AIS) compared with X-ray imaging due to its radiation-free superiority. Moreover, some deformities with low observability, such as vertebral rotation, in X-ray images can also be reflected by critical landmarks in 3D ultrasound data. In this paper, we propose a localization network (LLNet) to extract lamina in 3D ultrasound data, which has been indicated as a meaningful anatomy for measuring vertebral rotation by clinical studies. First, the LLNet skillfully establishes a parallel anatomical prior embedding branch that implicitly explores the anatomical correlation between the lamina and another anatomy with more stable observability (spinous process) during the training phase and then introduces the correlation to highlight the potential region of the lamina in the inferring one. Second, since the lamina is a tiny target, the information loss caused by continuous convolutional and pooling operations has a profound negative effect on detecting the lamina. We employ an optimization mechanism to mitigate this problem, which refines feature maps according to information from the original image and further reuses them to polish output. Furthermore, a modified global-local attention module is deployed on skip connections to mine global dependencies and contextual information to construct an effective image pattern. Extensive comparisons and ablation studies are performed on actual clinical data. Results indicate that the capability of our model is better than other outstanding detection models, and functional modules effectively contribute to this, with a 100.0 % detection success rate and an 8.9 % improvement of mean intersection over the union. Thus, our model is promising to become a helpful part of a computer-ass

关键词： Ultrasonic imaging

来源：评论

学校读者我要写书评

暂无评论

Decoupled deep hough voting for point cloud registration

引用

Frontiers of computer Science 2024年第2期18卷 147-155页

作者： Mingzhi YUAN Kexue FU Zhihao LI Manning WANG Digital Medical Research Center School of Basic Medical SciencesFudan UniversityShanghai 200032China Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention Shanghai 200032China

Estimating rigid transformation using noisy correspondences is critical to feature-based point cloud ***,a series of studies have attempted to combine traditional robust model fitting with deep *** them,DHVR proposed a hough voting-based method,achieving new state-of-the-art ***,we find voting on rotation and translation simultaneously hinders achieving better ***,we proposed a new hough voting-based method,which decouples rotation and translation ***,we first utilize hough voting and a neural network to estimate *** based on good initialization on rotation,we can easily obtain accurate rigid *** experiments on 3DMatch and 3DLoMatch datasets show that our method achieves comparable performances over the state-of-the-art *** further demonstrate the generalization of our method by experimenting on KITTI dataset.

关键词： point cloud registration robust model fitting deep learning hough voting

来源：评论

学校读者我要写书评

暂无评论

XSleepFusion: A dual-stage information bottleneck fusion framework for interpretable multimodal sleep analysis

引用

Information Fusion 2025年 123卷

作者： Shuaicong Hu Yanan Wang Jian Liu Cuiwei Yang Department of Biomedical Engineering Fudan University Shanghai 200433 China Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention of Shanghai Shanghai 200093 China

Sleep disorders affect hundreds of millions globally, with accurate assessment of sleep apnea (SA) and sleep staging (SS) essential for clinical diagnosis and early intervention. Manual analysis by sleep experts is time-consuming and subject to inter-rater variability. Deep learning (DL) approaches offer automation potential but face fundamental challenges in multi-modal physiological signal integration and interpretability. This paper presents XSleepFusion, a cross-modal fusion framework based on information bottleneck (IB) theory for automated sleep analysis. The framework introduces a dual-stage IB mechanism that systematically processes physiological signals: first eliminating intra-modal redundancy, then optimizing cross-modal feature fusion. An evolutionary attention Transformer network (EAT-Net) backbone extracts temporal features at multiple scales, providing interpretable attention patterns. Experimental validation on eight clinical datasets comprising over 15,000 sleep recordings demonstrates the framework’s effectiveness in polysomnogram (PSG)-based SA detection, electrocariogram (ECG)-based SA detection, and SS. The architecture achieves superior generalization across varying signal qualities and modal combinations, while the dual-stage design enables flexible integration of diverse physiological signals. Through interpretable feature representations and robust cross-modal fusion capabilities, XSleepFusion establishes a reliable and adaptable foundation for clinical sleep monitoring.

关键词：

来源：评论

学校读者我要写书评

暂无评论

PLTN: Noisy Label Learning in Long-Tailed medical images with Adaptive Prototypes

引用

Neurocomputing 2025年

作者： Zhiqing He Jie Xu Peng Huang Qin Wang Yuanyuan Wang Yi Guo Department of Biomedical Engineering School of Information Science and Technology Fudan University Shanghai China Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention of Shanghai Shanghai China

Label noise poses a significant challenge in medical image classification, particularly for minority diseases in long-tailed distribution. Most Noisy Label Learning methods rely on clean sample selection to learn representative patterns. However, the imbalanced data distribution and intra-class variability cause model bias towards head classes at the inter-class level and dominant patterns at the intra-class level. Existing methods show limited performance in handling these biases simultaneously, causing inefficient clean sample selection and hindering minority class training. Thus, we propose PLTN, named P rototype-guided L ong- T ail N oisy Learning, a novel approach that selects comprehensive clean samples from each class and enhances minority class training to improve diagnostic accuracy of rare disease in long tailed medical data. First, Adaptive Multi-Prototype Learning (AMPL) module is designed to enable precise selection of clean samples, especially for minority classes. It achieves this by capturing diverse feature patterns through a dynamic prototype learning approach. The clean samples are then treated as labeled data while the noisy samples are used as unlabeled data in a semi-supervised learning paradigm. Second, we introduce Sharpness-aware Distributional Mixup (SDM) to enhance minority class training. SDM employs distributional mixup to enrich tail classes representations, simultaneously minimizing loss landscape sharpness to enhance model generalization for all classes, benefiting tail classes. Experiments on three typical medical datasets demonstrate that our method outperforms state-of-the-art methods. On HAM10000 dataset with 20% label noise, we achieve an 80.34% Macro-F1 score, with notable gains in tail class performance while maintaining head class performance, showing potential for real-world clinical applications.

关键词： Long tail noisy learning medical image classification Adaptive multi-prototype learning Sharpness-aware distributional mixup

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：