检索结果-内蒙古大学图书馆

Dual Attention Poser: Dual Path Body Tracking Based on Attention

学校读者我要写书评

暂无评论

Dual Attention Poser: Dual Path Body Tracking Based on Atten...

IEEE computer Society Conference on computer Vision and Pattern Recognition Workshops (CVPRW)

作者： Xinhan Di Xiaokun Dai Xinkang Zhang Xinrong Chen Deepearthgo Academy for Engineering&Technology Fudan Universiry Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention Fudan University

Currently, mixed reality head-mounted displays tracking the full body of users is an important human-computer interaction mode through the pose of the head and the hands. Unfortunately, users’ virtual representation and experience is limited due to high reconstruction error when simple transformer network architecture is applied. In this paper, we present a novel model, named Dual Attention Poser, which can learn the whole body reconstruction at a high accuracy. The proposed model consists of three key modules. Among them, dual-path attention encoder is designed to extract feature of the sparse signals. Cross attention mixer module enable the fusion of representation in the double path. Attention-gated-mlp decoder is applied to decode the hidden feature from the sparse input through attention gate. Test results on the AMASS dataset show that Dual Attention Poser can reduce the error by up to 18.2% in comparison with the state-of-the-art results.

关键词：

FANCL: Feature-Guided Attention Network with Curriculum Learning for Brain Metastases Segmentation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Zijiang Liu, Xiaoyu Qu, Linhao Shi, Yonghong Digital Medical Research Center School of Basic Medical Science Fudan University Shanghai200032 China Shanghai Key Lab of Medical Image Computing and Computer Assisted Intervention Shanghai200032 China

Accurate segmentation of brain metastases (BMs) in MR image is crucial for the diagnosis and followup of patients. Methods based on deep convolutional neural networks (CNNs) have achieved high segmentation performance. However, due to the loss of critical feature information caused by convolutional and pooling operations, CNNs still face great challenges in small BMs segmentation. Besides, BMs are irregular and easily confused with healthy tissues, which makes it difficult for the model to effectively learn tumor structure during training. To address these issues, this paper proposes a novel model called feature-guided attention network with curriculum learning (FANCL). Based on CNNs, FANCL utilizes the input image and its feature to establish the intrinsic connections between metastases of different sizes, which can effectively compensate for the loss of high-level feature from small tumors with the information of large tumors. Furthermore, FANCL applies the voxel-level curriculum learning strategy to help the model gradually learn the structure and details of BMs. And baseline models of varying depths are employed as curriculum-mining networks for organizing the curriculum progression. The evaluation results on the BraTS-METS 2023 dataset indicate that FANCL significantly improves the segmentation performance, confirming the effectiveness of our method. Copyright © 2024, The Authors. All rights reserved.

关键词： Curricula

Deep Mutual Learning among Partially Labeled Datasets for Multi-Organ Segmentation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Xiaoyu Qu, Linhao Xie, Ziyue Shi, Yonghong Song, Zhijian Digital Medical Research Center School of Basic Medical Science Fudan University Shanghai200032 China Shanghai Key Lab of Medical Image Computing and Computer Assisted Intervention Shanghai200032 China

The task of labeling multiple organs for segmentation is a complex and time-consuming process, resulting in a scarcity of comprehensively labeled multi-organ datasets while the emergence of numerous partially labeled datasets. Current methods are inadequate in effectively utilizing the supervised information available from these datasets, thereby impeding the progress in improving the segmentation accuracy. This paper proposes a two-stage multi-organ segmentation method based on mutual learning, aiming to improve multi-organ segmentation performance by complementing information among partially labeled datasets. In the first stage, each partial-organ segmentation model utilizes the non-overlapping organ labels from different datasets and the distinct organ features extracted by different models, introducing additional mutual difference learning to generate higher quality pseudo labels for unlabeled organs. In the second stage, each full-organ segmentation model is supervised by fully labeled datasets with pseudo labels and leverages true labels from other datasets, while dynamically sharing accurate features across different models, introducing additional mutual similarity learning to enhance multi-organ segmentation performance. Extensive experiments were conducted on nine datasets that included the head and neck, chest, abdomen, and pelvis. The results indicate that our method has achieved SOTA performance in segmentation tasks that rely on partial labels, and the ablation studies have thoroughly confirmed the efficacy of the mutual learning mechanism. © 2024, CC BY.

关键词： Deep learning

CHEST-DIFFUSION: A LIGHT-WEIGHT TEXT-TO-image MODEL FOR REPORT-TO-CXR GENERATION

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Huang, Peng Gao, Xue Huang, Lihong Jiao, Jing Li, Xiaokang Wang, Yuanyuan Guo, Yi Department of Electronic Engineering Fudan University Shanghai200433 China Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention of Shanghai Shanghai200032 China

Text-to-image generation has important implications for generation of diverse and controllable images. Several attempts have been made to adapt Stable Diffusion (SD) to the medical domain. However, the large distribution difference between medical reports and natural texts, as well as high computational complexity in common stable diffusion limit the authenticity and feasibility of the generated medical images. To solve above problems, we propose a novel lightweight transformer-based diffusion model learning framework, Chest-Diffusion, for report-to-CXR generation. Chest-Diffusion employs a domain-specific text encoder to obtain accurate and expressive text features to guide image generation, improving the authenticity of the generated images. Meanwhile, we introduce a light-weight transformer architecture as the denoising model, reducing the computational complexity of the diffusion model. Experiments demonstrate that our Chest-Diffusion achieves the lowest FID score 24.456, under the computation budget of 118.918 GFLOPs, which is nearly one-third of the computational complexity of SD. Copyright © 2024, The Authors. All rights reserved.

关键词： Computational complexity

Distinct immune escape and microenvironment between RG-like and pri-OPC-like glioma revealed by single-cell RNA-seq analysis

学校读者我要写书评

暂无评论

Frontiers of Medicine 2024年第1期18卷 147-168页

作者： Weiwei Xian Mohammad Asad Shuai Wu Zhixin Bai Fengjiao Li Junfeng Lu Gaoyu Zu Erin Brintnell Hong Chen Ying Mao Guomin Zhou Bo Liao Jinsong Wu Edwin Wang Linya You Department of Human Anatomy&Histoembryology School of Basic Medical SciencesFudan UniversityShanghai200032China Cumming School of Medicine University of CalgaryCalgaryAlbertaT2N 4N1Canada Glioma Surgery Division Department of NeurosurgeryHuashan HospitalFudan UniversityShanghai200040China Department of Pathology Huashan HospitalFudan UniversityShanghai200040China Department of Neurosurgery Huashan HospitalFudan UniversityShanghai200040China Shanghai Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention Fudan UniversityShanghai200040China School of Mathematics and Statistics Hainan Normal UniversityHaikou570100China

The association of neurogenesis and gliogenesis with glioma remains *** conducting single-cell RNA-seq analyses on 26 gliomas,we reported their classification into primitive oligodendrocyte precursor cell(pri-OPC)-like and radial glia(RG)-like tumors and validated it in a public cohort and TCGA *** RG-like tumors exhibited wild-type isocitrate dehydrogenase and tended to carry EGFR mutations,and the pri-OPC-like ones were prone to carrying TP53 *** subclones only in pri-OPC-like tumors showed substantially down-regulated MHC-I genes,suggesting their distinct immune evasion ***,the two subgroups appeared to extensively modulate glioma-infiltrating lymphocytes in distinct *** specific genes not expressed in normal immune cells were found in glioma-infiltrating *** example,glial/glioma stem cell markers OLIG1/PTPRZ1 and B cell-specific receptors IGLC2/IGKC were expressed in pri-OPC-like and RG-like glioma-infiltrating lymphocytes,*** expression was positively correlated with those of immune checkpoint genes(e.g.,LGALS33)and poor survivals as validated by the increased expression of LGALS3 upon IGKC overexpression in Jurkat *** finding indicated a potential inhibitory role in tumor-infiltrating lymphocytes and could provide a new way of cancer immune evasion.

关键词： single-cell RNA-seq glioma radial glia primitive oligodendrocyte precursor cell immune escape

Boosting ViT-based MRI Reconstruction from the Perspectives of Frequency Modulation, Spatial Purification, and Scale Diversification 39

学校读者我要写书评

暂无评论

Boosting ViT-based MRI Reconstruction from the Perspectives ...

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Meng, Yucong Yang, Zhiwei Shi, Yonghong Song, Zhijian 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 for Engineering and Technology Fudan University Shanghai200433 China

ISBN: (纸本)157735897X

The accelerated MRI reconstruction process presents a challenging ill-posed inverse problem due to the extensive undersampling in k-space. Recently, Vision Transformers (ViTs) have become the mainstream for this task, demonstrating substantial performance improvements. However, there are still three significant issues remain unaddressed: (1) ViTs struggle to capture high-frequency components of images, limiting their ability to detect local textures and edge information, thereby impeding MRI restoration;(2) Previous methods calculate multi-head self-attention (MSA) among both related and unrelated tokens in content, introducing noise and significantly increasing computational burden;(3) The naive feed-forward network in ViTs cannot model the multi-scale information that is important for image restoration. In this paper, we propose FPS-Former, a powerful ViT-based framework, to address these issues from the perspectives of frequency modulation, spatial purification, and scale diversification. Specifically, for issue (1), we introduce a frequency modulation attention module to enhance the self-attention map by adaptively re-calibrating the frequency information in a Laplacian pyramid. For issue (2), we customize a spatial purification attention module to capture interactions among closely related tokens, thereby reducing redundant or irrelevant feature representations. For issue (3), we propose an efficient feed-forward network based on a hybrid-scale fusion strategy. Comprehensive experiments conducted on three public datasets show that our FPS-Former outperforms state-of-the-art methods while requiring lower computational costs. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Frequency modulation

A Feasible Way to Assess the Lower Extremity Motor Function of Stroke Patients - A Preliminary Study 15

学校读者我要写书评

暂无评论

A Feasible Way to Assess the Lower Extremity Motor Function ...

15th International Congress on image and Signal Processing, Biomedical Engineering and Informatics, CISP-BMEI 2022

作者： Liu, Sen Xie, Hongyu Yuan, Han Tian, Jing Yang, Cuiwei Wu, Yi Center for Biomedical Engineering Fudan University Shanghai China Department of Rehabilitation Medicine Huashan Hospital Fudan University Shanghai China The Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention of Shanghai Shanghai China

ISBN: (数字)9781665488877

ISBN: (纸本)9781665488877

Accurate assessment of the mobility of stroke patients is crucial to their rehabilitation management. In order to quantify the lower extremity motor function in stroke sufferers, two tasks, toe tapping and knee flexion and extension, were designed based on the Brunnstrom approach. In this work, five hemiplegic patients in Brunnstrom stage III or IV were recruited. Acceleration signals under different tasks were recorded using tri-axial accelerometers, which were placed at the arches of the feet. Various kinematic parameters were then calculated, and a statistical test was conveyed to investigate the correlations between Brunnstrom stages and the features. The results showed that in the toe tapping task, the dominant frequency of the y-axis and the proportion of signal power in specific frequency band of the x-axis were significantly higher in stage III than in stage IV. On the contrary, the standard deviation and the skewness of the y-axis were lower in stage III. Besides, there were also significant differences in the amplitude of knee flexion between cohorts at different Brunnstrom stages. It is expected that the findings of this study would be helpful for the monitoring and rehabilitation of stroke patients. © 2022 IEEE.

关键词： Patient rehabilitation

SP3 : Superpixel-propagated pseudo-label learning for weakly semi-supervised medical image segmentation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Li, Shiman Zhao, Jiayue Liu, Shaolei Dai, Xiaokun Zhang, Chenxi Song, Zhijian The School of Basic Medical Science Fudan University 200433 China Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention 200032 China The Academy for Engineering and Technology Fudan University 200433 China

Deep learning-based medical image segmentation helps assist diagnosis and accelerate the treatment process while the model training usually requires large-scale dense annotation datasets. Weakly semi-supervised medical image segmentation is an essential application because it only requires a small amount of scribbles and a large number of unlabeled data to train the model, which greatly reduces the clinician’s effort to fully annotate images. To handle the inadequate supervisory information challenge in weakly semi-supervised segmentation (WSSS), a SuperPixel-Propagated Pseudo-label (SP3) learning method is proposed, using the structural information contained in superpixel for supplemental information. Specifically, the annotation of scribbles is propagated to superpixels and thus obtains a dense annotation for supervised training. Since the quality of pseudo-labels is limited by the low-quality annotation, the beneficial superpixels selected by dynamic thresholding are used to refine pseudo-labels. Furthermore, aiming to alleviate the negative impact of noise in pseudo-label, superpixel-level uncertainty is incorporated to guide the pseudo-label supervision for stable learning. Our method achieves state-of-the-art performance on both tumor and organ segmentation datasets under the WSSS setting, using only 3% of the annotation workload compared to fully supervised methods and attaining approximately 80% Dice score. Additionally, our method outperforms eight weakly and semi-supervised methods under both weakly supervised and semi-supervised settings. Results of extensive experiments validate the effectiveness and annotation efficiency of our weakly semi-supervised segmentation, which can assist clinicians in achieving automated segmentation for organs or tumors quickly and ultimately benefit patients. © 2024, CC BY.

关键词： Superpixels

The rise of AI language pathologists: exploring two-level prompt learning for few-shot weakly-supervised whole slide image classification 23

学校读者我要写书评

暂无评论

The rise of AI language pathologists: exploring two-level pr...

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Linhao Qu Xiaoyuan Luo Kexue Fu Manning Wang Zhijian Song Digital Medical Research Center School of Basic Medical Science Fudan University and Shanghai Key Lab of Medical Image Computing and Computer Assisted Intervention

This paper introduces the novel concept of few-shot weakly supervised learning for pathology Whole Slide image (WSI) classification, denoted as FSWC. A solution is proposed based on prompt learning and the utilization of a large language model, GPT-4. Since a WSI is too large and needs to be divided into patches for processing, WSI classification is commonly approached as a Multiple Instance Learning (MIL) problem. In this context, each WSI is considered a bag, and the obtained patches are treated as instances. The objective of FSWC is to classify both bags and instances with only a limited number of labeled bags. Unlike conventional few-shot learning problems, FSWC poses additional challenges due to its weak bag labels within the MIL framework. Drawing inspiration from the recent achievements of vision-language models (V-L models) in downstream few-shot classification tasks, we propose a two-level prompt learning MIL framework tailored for pathology, incorporating language prior knowledge. Specifically, we leverage CLIP to extract instance features for each patch, and introduce a prompt-guided pooling strategy to aggregate these instance features into a bag feature. Subsequently, we employ a small number of labeled bags to facilitate few-shot prompt learning based on the bag features. Our approach incorporates the utilization of GPT-4 in a question-and-answer mode to obtain language prior knowledge at both the instance and bag levels, which are then integrated into the instance and bag level language prompts. Additionally, a learnable component of the language prompts is trained using the available few-shot labeled data. We conduct extensive experiments on three real WSI datasets encompassing breast cancer, lung cancer, and cervical cancer, demonstrating the notable performance of the proposed method in bag and instance classification. Codes will be available at https://***/miccaiif/TOP.

关键词：