IEEE transactions on medical imaging最新文献_第10页

VibNet: Vibration-Boosted Needle Detection in Ultrasound Images VibNet：超声图像中的振动增强针检测

IEEE transactions on medical imaging Pub Date : 2025-02-25 DOI: 10.1109/TMI.2025.3545434

Dianye Huang;Chenyang Li;Angelos Karlas;Xiangyu Chu;K. W. Samuel Au;Nassir Navab;Zhongliang Jiang

{"title":"VibNet: Vibration-Boosted Needle Detection in Ultrasound Images","authors":"Dianye Huang;Chenyang Li;Angelos Karlas;Xiangyu Chu;K. W. Samuel Au;Nassir Navab;Zhongliang Jiang","doi":"10.1109/TMI.2025.3545434","DOIUrl":"10.1109/TMI.2025.3545434","url":null,"abstract":"Precise percutaneous needle detection is crucial for ultrasound (US)-guided interventions. However, inherent limitations such as speckles, needle-like artifacts, and low resolution make it challenging to robustly detect needles, especially when their visibility is reduced or imperceptible. To address this challenge, we propose VibNet, a learning-based framework designed to enhance the robustness and accuracy of needle detection in US images by leveraging periodic vibration applied externally to the needle shafts. VibNet integrates neural Short-Time Fourier Transform and Hough Transform modules to achieve successive sub-goals, including motion feature extraction in the spatiotemporal space, frequency feature aggregation, and needle detection in the Hough space. Due to the periodic subtle vibration, the features are more robust in the frequency domain than in the image intensity domain, making VibNet more effective than traditional intensity-based methods. To demonstrate the effectiveness of VibNet, we conducted experiments on distinct ex vivo porcine and bovine tissue samples. The results obtained on porcine samples demonstrate that VibNet effectively detects needles even when their visibility is severely reduced, with a tip error of <inline-formula> <tex-math>${1}.{61}pm {1}.{56}~textit {mm}$ </tex-math></inline-formula> compared to <inline-formula> <tex-math>${8}.{15}pm {9}.{98}~textit {mm}$ </tex-math></inline-formula> for UNet and <inline-formula> <tex-math>${6}.{63}pm {7}.{58}~textit {mm}$ </tex-math></inline-formula> for WNet, and a needle direction error of <inline-formula> <tex-math>${1}.{64}pm {1}.{86}^{circ }$ </tex-math></inline-formula> compared to <inline-formula> <tex-math>${9}.{29}~pm ~{15}.{30}^{circ }$ </tex-math></inline-formula> for UNet and <inline-formula> <tex-math>${8}.{54}~pm ~{17}.{92}^{circ }$ </tex-math></inline-formula> for WNet. Code: <uri>https://github.com/marslicy/VibNet</uri>.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 6","pages":"2696-2708"},"PeriodicalIF":0.0,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10902567","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143495369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Foundation Model-Guided Gaussian Splatting for 4D Reconstruction of Deformable Tissues 变形组织四维重建的基础模型引导高斯溅射

IEEE transactions on medical imaging Pub Date : 2025-02-25 DOI: 10.1109/TMI.2025.3545183

Yifan Liu;Chenxin Li;Hengyu Liu;Chen Yang;Yixuan Yuan

{"title":"Foundation Model-Guided Gaussian Splatting for 4D Reconstruction of Deformable Tissues","authors":"Yifan Liu;Chenxin Li;Hengyu Liu;Chen Yang;Yixuan Yuan","doi":"10.1109/TMI.2025.3545183","DOIUrl":"10.1109/TMI.2025.3545183","url":null,"abstract":"Reconstructing deformable anatomical structures from endoscopic videos is a pivotal and promising research topic that can enable advanced surgical applications and improve patient outcomes. While existing surgical scene reconstruction methods have made notable progress, they often suffer from slow rendering speeds due to using neural radiance fields, limiting their practical viability in real-world applications. To overcome this bottleneck, we propose EndoGaussian, a framework that integrates the strengths of 3D Gaussian Splatting representations, allowing for high-fidelity tissue reconstruction, efficient training, and real-time rendering. Specifically, we dedicate a Foundation Model-driven Initialization (FMI) module, which distills 3D cues from multiple vision foundation models (VFMs) to swiftly construct the preliminary scene structure for Gaussian initialization. Then, a Spatio-temporal Gaussian Tracking (SGT) is designed, efficiently modeling scene dynamics using the multi-scale HexPlane with spatio-temporal priors. Furthermore, to improve the dynamics modeling ability for scenes with large deformation, EndoGaussian integrates Motion-aware Frame Synthesis (MFS) to adaptively synthesize new frames as extra training constraints. Experimental results on public datasets demonstrate EndoGaussian’s efficacy against prior state-of-the-art methods, including superior rendering speed (168 FPS, real-time), enhanced rendering quality (38.555 PSNR), and reduced training overhead (within 2 min/scene). These results underscore EndoGaussian’s potential to significantly advance intraoperative surgery applications, paving the way for more accurate and efficient real-time surgical guidance and decision-making in clinical scenarios. Code is available at: <uri>https://github.com/CUHK-AIM-Group/EndoGaussian</uri>.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 6","pages":"2672-2682"},"PeriodicalIF":0.0,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143495368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Dual-Scanning Photoacoustic Endomicroscopy for High-Speed Gastrointestinal Microvascular Imaging 双扫描光声内镜用于高速胃肠道微血管成像

IEEE transactions on medical imaging Pub Date : 2025-02-21 DOI: 10.1109/TMI.2025.3544403

Hongdian Sun;Xiao Liang;Linyang Li;Yuanlong Zhao;Heng Guo;Weizhi Qi;Lei Xi

引用次数: 0

Organ-DETR: Organ Detection via Transformers 器官- detr：通过变压器进行器官检测

IEEE transactions on medical imaging Pub Date : 2025-02-19 DOI: 10.1109/TMI.2025.3543581

Morteza Ghahremani;Benjamin Raphael Ernhofer;Jiajun Wang;Marcus Makowski;Christian Wachinger

{"title":"Organ-DETR: Organ Detection via Transformers","authors":"Morteza Ghahremani;Benjamin Raphael Ernhofer;Jiajun Wang;Marcus Makowski;Christian Wachinger","doi":"10.1109/TMI.2025.3543581","DOIUrl":"10.1109/TMI.2025.3543581","url":null,"abstract":"Query-based Transformers have been yielding impressive performance in object localization and detection tasks. However, their application to organ detection in 3D medical imaging data has been relatively unexplored. This study introduces Organ-DETR, featuring two innovative modules, MultiScale Attention (MSA) and Dense Query Matching (DQM), designed to enhance the performance of Detection Transformers (DETRs) for 3D organ detection. MSA is a novel top-down representation learning approach for efficiently encoding Computed Tomography (CT) features. This architecture employs a multiscale attention mechanism, utilizing both dual self-attention and cross-scale attention mechanisms to extract intra- and inter-scale spatial interactions in the attention mechanism. Organ-DETR also introduces DQM, an approach for one-to-many matching that tackles the label assignment difficulties in organ detection. DQM increases positive queries to enhance both recall scores and training efficiency without the need for additional learnable parameters. Extensive results on five 3D CT datasets indicate that the proposed Organ-DETR outperforms comparable techniques by achieving a remarkable improvement of +10.6 mAP COCO. The project and code are available at <uri>https://github.com/ai-med/OrganDETR</uri>.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 6","pages":"2657-2671"},"PeriodicalIF":0.0,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10892276","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143452341","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Editorial Flagship Toward the Future 面向未来的旗舰

IEEE transactions on medical imaging Pub Date : 2025-02-17 DOI: 10.1109/TMI.2025.3543049

Ge Wang

引用次数: 0

Cross- and Intra-Image Prototypical Learning for Multi-Label Disease Diagnosis and Interpretation 用于多标签疾病诊断和解释的交叉和图像内原型学习

IEEE transactions on medical imaging Pub Date : 2025-02-14 DOI: 10.1109/TMI.2025.3541830

Chong Wang;Fengbei Liu;Yuanhong Chen;Helen Frazer;Gustavo Carneiro

{"title":"Cross- and Intra-Image Prototypical Learning for Multi-Label Disease Diagnosis and Interpretation","authors":"Chong Wang;Fengbei Liu;Yuanhong Chen;Helen Frazer;Gustavo Carneiro","doi":"10.1109/TMI.2025.3541830","DOIUrl":"10.1109/TMI.2025.3541830","url":null,"abstract":"Recent advances in prototypical learning have shown remarkable potential to provide useful decision interpretations associating activation maps and predictions with class-specific training prototypes. Such prototypical learning has been well-studied for various single-label diseases, but for quite relevant and more challenging multi-label diagnosis, where multiple diseases are often concurrent within an image, existing prototypical learning models struggle to obtain meaningful activation maps and effective class prototypes due to the entanglement of the multiple diseases. In this paper, we present a novel Cross- and Intra-image Prototypical Learning (CIPL) framework, for accurate multi-label disease diagnosis and interpretation from medical images. CIPL takes advantage of common cross-image semantics to disentangle the multiple diseases when learning the prototypes, allowing a comprehensive understanding of complicated pathological lesions. Furthermore, we propose a new two-level alignment-based regularisation strategy that effectively leverages consistent intra-image information to enhance interpretation robustness and predictive performance. Extensive experiments show that our CIPL attains the state-of-the-art (SOTA) classification accuracy in two public multi-label benchmarks of disease diagnosis: thoracic radiography and fundus images. Quantitative interpretability results show that CIPL also has superiority in weakly-supervised thoracic disease localisation over other leading saliency- and prototype-based explanation methods.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 6","pages":"2568-2580"},"PeriodicalIF":0.0,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143417404","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

TransMatch: Employing Bridging Strategy to Overcome Large Deformation for Feature Matching in Gastroscopy Scenario TransMatch：利用桥接策略克服胃镜场景中大变形的特征匹配

IEEE transactions on medical imaging Pub Date : 2025-02-13 DOI: 10.1109/TMI.2025.3541433

Guosong Zhu;Zhen Qin;Linfang Yu;Yi Ding;Zhiguang Qin

引用次数: 0

Method for Correcting the Muscle Fiber Orientation Determined by a T-Shaped Transducer in Ultrasound Shear Wave Elastography 超声横波弹性成像中t形传感器测定肌纤维方向的校正方法

IEEE transactions on medical imaging Pub Date : 2025-02-13 DOI: 10.1109/TMI.2025.3541321

Chien Chen;Guo-Xuan Xu;Wei-Ren Su;Chih-Chung Huang

{"title":"Method for Correcting the Muscle Fiber Orientation Determined by a T-Shaped Transducer in Ultrasound Shear Wave Elastography","authors":"Chien Chen;Guo-Xuan Xu;Wei-Ren Su;Chih-Chung Huang","doi":"10.1109/TMI.2025.3541321","DOIUrl":"10.1109/TMI.2025.3541321","url":null,"abstract":"Shear wave elastography (SWE) is a quantitative imaging method that could be used for clinical assessment of musculoskeletal stiffness, particularly in disease diagnosis and rehabilitation evaluation. However, the elastic anisotropy of skeletal muscle leads to uncertainties in shear wave velocity (SWV) measurements in SWE because the SWV varies with muscle fiber orientation. Therefore, many studies have conducted 360° rotational measurements of SWV to determine the elastic anisotropy of muscle; however, the extended data acquisition time of this approach limits its clinical utility. In this study, a T-shaped transducer was used for rapidly measuring the longitudinal and transverse SWVs (<inline-formula> <tex-math>$textit {SWV}_{L}$ </tex-math></inline-formula> and <inline-formula> <tex-math>$textit {SWV}_{T}$ </tex-math></inline-formula>) of muscle through an ellipse fitting method to estimate the fiber orientation angle when the excitation is normal to the material axis. The performance of this approach was examined by conducting a homogeneous elastic phantom experiment, which indicated that the proposed T-shaped transducer generated shear waves in three directions by applying a supersonic push at the junction of the transducer. The error between the measured SWVs and ground truth was approximately 6.5%. The proposed T-shaped transducer was also used to measure the SWV in the biceps brachii of four healthy individuals. The <inline-formula> <tex-math>$textit {SWV}_{L}$ </tex-math></inline-formula> and <inline-formula> <tex-math>$textit {SWV}_{T}$ </tex-math></inline-formula> values measured with this transducer were 2.47 and 1.09 m/s, respectively, which were consistent with the SWVs obtained under 360° rotation and in the literature (an error of ~4%). All experimental results were consistent with the results obtained under 360° rotation, which indicates that the proposed method enables the rapid and stable estimation of muscle fiber orientation in SWE.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 6","pages":"2528-2540"},"PeriodicalIF":0.0,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143417409","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

CT-SDM: A Sampling Diffusion Model for Sparse-View CT Reconstruction Across Various Sampling Rates CT- sdm：不同采样率下稀疏视图CT重建的采样扩散模型

IEEE transactions on medical imaging Pub Date : 2025-02-13 DOI: 10.1109/TMI.2025.3541491

Liutao Yang;Jiahao Huang;Guang Yang;Daoqiang Zhang

{"title":"CT-SDM: A Sampling Diffusion Model for Sparse-View CT Reconstruction Across Various Sampling Rates","authors":"Liutao Yang;Jiahao Huang;Guang Yang;Daoqiang Zhang","doi":"10.1109/TMI.2025.3541491","DOIUrl":"10.1109/TMI.2025.3541491","url":null,"abstract":"Sparse views X-ray computed tomography has emerged as a contemporary technique to mitigate radiation dose. Because of the reduced number of projection views, traditional reconstruction methods can lead to severe artifacts. Recently, research studies utilizing deep learning methods has made promising progress in removing artifacts for Sparse-View Computed Tomography (SVCT). However, given the limitations on the generalization capability of deep learning models, current methods usually train models on fixed sampling rates, affecting the usability and flexibility of model deployment in real clinical settings. To address this issue, our study proposes a adaptive reconstruction method to achieve high-performance SVCT reconstruction at various sampling rate. Specifically, we design a novel imaging degradation operator in the proposed sampling diffusion model for SVCT (CT-SDM) to simulate the projection process in the sinogram domain. Thus, the CT-SDM can gradually add projection views to highly undersampled measurements to generalize the full-view sinograms. By choosing an appropriate starting point in diffusion inference, the proposed model can recover the full-view sinograms from various sampling rate with only one trained model. Experiments on several datasets have verified the effectiveness and robustness of our approach, demonstrating its superiority in reconstructing high-quality images from sparse-view CT scans across various sampling rates.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 6","pages":"2581-2593"},"PeriodicalIF":0.0,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143417407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Masked Vascular Structure Segmentation and Completion in Retinal Images 视网膜图像中蒙膜血管结构的分割与补全

IEEE transactions on medical imaging Pub Date : 2025-02-13 DOI: 10.1109/TMI.2025.3538336

Yi Zhou;Thiara Sana Ahmed;Meng Wang;Eric A. Newman;Leopold Schmetterer;Huazhu Fu;Jun Cheng;Bingyao Tan

{"title":"Masked Vascular Structure Segmentation and Completion in Retinal Images","authors":"Yi Zhou;Thiara Sana Ahmed;Meng Wang;Eric A. Newman;Leopold Schmetterer;Huazhu Fu;Jun Cheng;Bingyao Tan","doi":"10.1109/TMI.2025.3538336","DOIUrl":"10.1109/TMI.2025.3538336","url":null,"abstract":"Early retinal vascular changes in diseases such as diabetic retinopathy often occur at a microscopic level. Accurate evaluation of retinal vascular networks at a micro-level could significantly improve our understanding of angiopathology and potentially aid ophthalmologists in disease assessment and management. Multiple angiogram-related retinal imaging modalities, including fundus, optical coherence tomography angiography, and fluorescence angiography, project continuous, inter-connected retinal microvascular networks into imaging domains. However, extracting the microvascular network, which includes arterioles, venules, and capillaries, is challenging due to the limited contrast and resolution. As a result, the vascular network often appears as fragmented segments. In this paper, we propose a backbone-agnostic Masked Vascular Structure Segmentation and Completion (MaskVSC) method to reconstruct the retinal vascular network. MaskVSC simulates missing sections of blood vessels and uses this simulation to train the model to predict the missing parts and their connections. This approach simulates highly heterogeneous forms of vessel breaks and mitigates the need for massive data labeling. Accordingly, we introduce a connectivity loss function that penalizes interruptions in the vascular network. Our findings show that masking 40% of the segments yields optimal performance in reconstructing the interconnected vascular network. We test our method on three different types of retinal images across five separate datasets. The results demonstrate that MaskVSC outperforms state-of-the-art methods in maintaining vascular network completeness and segmentation accuracy. Furthermore, MaskVSC has been introduced to different segmentation backbones and has successfully improved performance. The code and 2PFM data are available at: <uri>https://github.com/Zhouyi-Zura/MaskVSC</uri>.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 6","pages":"2492-2503"},"PeriodicalIF":0.0,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10887048","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143417412","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0