The right atrium (RA) is critical for cardiac hemodynamics but is often overlooked in clinical diagnostics. This study presents a benchmark framework for RA cavity segmentation from late gadolinium-enhanced magnetic resonance imaging (LGE-MRIs), leveraging a two-stage strategy and a novel 3D deep learning network, RASnet. The architecture addresses challenges in class imbalance and anatomical variability by incorporating multi-path input, multi-scale feature fusion modules, Vision Transformers, context interaction mechanisms, and deep supervision. Evaluated on datasets comprising 354 LGE-MRIs, RASnet achieves SOTA performance with a Dice score of 92.19% on a primary dataset and demonstrates robust generalizability on an independent dataset. The proposed framework establishes a benchmark for RA cavity segmentation, enabling accurate and efficient analysis for cardiac imaging applications. Open-source code (https://github.com/zjinw/RAS) and data (https://zenodo.org/records/15524472) are provided to facilitate further research and clinical adoption.
{"title":"A Benchmark Framework for the Right Atrium Cavity Segmentation From LGE-MRIs","authors":"Jieyun Bai;Jinwen Zhu;Zhiting Chen;Ziduo Yang;Yaosheng Lu;Lei Li;Qince Li;Wei Wang;Henggui Zhang;Kuanquan Wang;Jie Gan;Jichao Zhao;Hua Lu;Suining Li;Jiawen Huang;Xiaoming Chen;Xiaoshen Zhang;Xiaowei Xu;Lulu Li;Yanfeng Tian;Víctor M. Campello;Karim Lekadir","doi":"10.1109/TMI.2025.3590694","DOIUrl":"10.1109/TMI.2025.3590694","url":null,"abstract":"The right atrium (RA) is critical for cardiac hemodynamics but is often overlooked in clinical diagnostics. This study presents a benchmark framework for RA cavity segmentation from late gadolinium-enhanced magnetic resonance imaging (LGE-MRIs), leveraging a two-stage strategy and a novel 3D deep learning network, RASnet. The architecture addresses challenges in class imbalance and anatomical variability by incorporating multi-path input, multi-scale feature fusion modules, Vision Transformers, context interaction mechanisms, and deep supervision. Evaluated on datasets comprising 354 LGE-MRIs, RASnet achieves SOTA performance with a Dice score of 92.19% on a primary dataset and demonstrates robust generalizability on an independent dataset. The proposed framework establishes a benchmark for RA cavity segmentation, enabling accurate and efficient analysis for cardiac imaging applications. Open-source code (<uri>https://github.com/zjinw/RAS</uri>) and data (<uri>https://zenodo.org/records/15524472</uri>) are provided to facilitate further research and clinical adoption.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 12","pages":"5290-5305"},"PeriodicalIF":0.0,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144684346","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}
Surgical action triplet detection offers intuitive intraoperative scene analysis for dynamically perceiving laparoscopic surgical workflows and analyzing the interaction between instruments and tissues. The current challenge of this task lies in simultaneously localizing surgical instruments while performing more accurate surgical triplet recognition to enhance a comprehensive understanding of intraoperative surgical scenes. To fully leverage the spatial localization of surgical instruments for associating with triplet detection, we propose an Instrument-Tissue-Guided Triplet detector, termed ITG-Trip, which navigates the confluence of surgical action cues through instrument and tissue pseudo-localization labeling to optimize action triplet detection. For exploiting textual and temporal trails, our framework embraces a Visual-Linguistic Association (VLA) module that exploits a pre-trained text encoder to distill textual prior knowledge, enhancing semantic information in global visual features and compensating rare interaction class perception. Besides, we introduce a Mamba-enhanced Spatial-temporal Perception (MSP) decoder, which weaves Mamba and Transformer blocks to explore subject- and object-aware spatial and temporal information to improve the accuracy of action triplet detection in long-time sequence surgical videos. Experimental results on the CholecT50 benchmark indicate that our method significantly outperforms existing state-of-the-art methods in both instrument localization and action triplet detection. The code is available at: github.com/PJLallen/ITG-Trip
{"title":"Instrument-Tissue-Guided Surgical Action Triplet Detection via Textual-Temporal Trail Exploration","authors":"Jialun Pei;Jiaan Zhang;Guanyi Qin;Kai Wang;Yueming Jin;Pheng-Ann Heng","doi":"10.1109/TMI.2025.3590457","DOIUrl":"10.1109/TMI.2025.3590457","url":null,"abstract":"Surgical action triplet detection offers intuitive intraoperative scene analysis for dynamically perceiving laparoscopic surgical workflows and analyzing the interaction between instruments and tissues. The current challenge of this task lies in simultaneously localizing surgical instruments while performing more accurate surgical triplet recognition to enhance a comprehensive understanding of intraoperative surgical scenes. To fully leverage the spatial localization of surgical instruments for associating with triplet detection, we propose an Instrument-Tissue-Guided Triplet detector, termed ITG-Trip, which navigates the confluence of surgical action cues through instrument and tissue pseudo-localization labeling to optimize action triplet detection. For exploiting textual and temporal trails, our framework embraces a Visual-Linguistic Association (VLA) module that exploits a pre-trained text encoder to distill textual prior knowledge, enhancing semantic information in global visual features and compensating rare interaction class perception. Besides, we introduce a Mamba-enhanced Spatial-temporal Perception (MSP) decoder, which weaves Mamba and Transformer blocks to explore subject- and object-aware spatial and temporal information to improve the accuracy of action triplet detection in long-time sequence surgical videos. Experimental results on the CholecT50 benchmark indicate that our method significantly outperforms existing state-of-the-art methods in both instrument localization and action triplet detection. The code is available at: github.com/PJLallen/ITG-Trip","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 12","pages":"5278-5289"},"PeriodicalIF":0.0,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144661828","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}
Pub Date : 2025-07-18DOI: 10.1109/TMI.2025.3590484
Pengpeng Sheng;Gangming Zhao;Tingting Han;Lei Qu
Effective representation of neuronal morphology is essential for cell typing and understanding brain function. However, the complexity of neuronal morphology manifests not only in inter-class structural differences but also in intra-class variations across developmental stages and environmental conditions. Such diversity poses significant challenges for existing methods in balancing robustness and discriminative power when representing neuronal morphology. To address this, we propose SGTMorph, a hybrid Graph Transformer framework that leverages the local topological modeling capabilities of graph neural networks and the global relational reasoning strengths of Transformers to explicitly encode neuronal structural information. SGTMorph incorporates a random walk-based positional encoding scheme to facilitate effective information propagation across neuronal graphs and introduces a spatially invariant encoding mechanism to improve adaptability with diverse morphology. This integrated approach enables a robust and comprehensive representation of neuronal morphology while maintaining biological fidelity. To enable label-free feature learning, we devise a self-supervised learning strategy grounded in geometric and topological similarity metrics. Extensive experiments on five datasets demonstrate SGTMorph’s superior performance in neuron morphology classification and retrieval tasks. Furthermore, Its practical utility in neuronal function research is validated through the accurate predictions of two functional features: the laminar distribution of somas and axonal projection patterns. The code is available at https://github.com/big-rain/SGTMorph
{"title":"Self-Supervised Neuron Morphology Representation With Graph Transformer","authors":"Pengpeng Sheng;Gangming Zhao;Tingting Han;Lei Qu","doi":"10.1109/TMI.2025.3590484","DOIUrl":"10.1109/TMI.2025.3590484","url":null,"abstract":"Effective representation of neuronal morphology is essential for cell typing and understanding brain function. However, the complexity of neuronal morphology manifests not only in inter-class structural differences but also in intra-class variations across developmental stages and environmental conditions. Such diversity poses significant challenges for existing methods in balancing robustness and discriminative power when representing neuronal morphology. To address this, we propose SGTMorph, a hybrid Graph Transformer framework that leverages the local topological modeling capabilities of graph neural networks and the global relational reasoning strengths of Transformers to explicitly encode neuronal structural information. SGTMorph incorporates a random walk-based positional encoding scheme to facilitate effective information propagation across neuronal graphs and introduces a spatially invariant encoding mechanism to improve adaptability with diverse morphology. This integrated approach enables a robust and comprehensive representation of neuronal morphology while maintaining biological fidelity. To enable label-free feature learning, we devise a self-supervised learning strategy grounded in geometric and topological similarity metrics. Extensive experiments on five datasets demonstrate SGTMorph’s superior performance in neuron morphology classification and retrieval tasks. Furthermore, Its practical utility in neuronal function research is validated through the accurate predictions of two functional features: the laminar distribution of somas and axonal projection patterns. The code is available at <uri>https://github.com/big-rain/SGTMorph</uri>","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 12","pages":"5332-5344"},"PeriodicalIF":0.0,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144661826","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}
Pub Date : 2025-07-16DOI: 10.1109/TMI.2025.3589928
Fuying Wang;Lequan Yu
Foundation models have significantly revolutionized the field of chest X-ray diagnosis with their ability to transfer across various diseases and tasks. However, previous works have predominantly utilized self-supervised learning from medical image-text pairs, which falls short in dense medical prediction tasks due to their sole reliance on such coarse pair supervision, thereby limiting their applicability to detailed diagnostics. In this paper, we introduce a Dense Chest X-ray Foundation Model (DCXFM), which utilizes mixed supervision types (i.e., text, label, and segmentation masks) to significantly enhance the scalability of foundation models across various medical tasks. Our model involves two training stages: we first employ a novel self-distilled multimodal pretraining paradigm to exploit text and label supervision, along with local-to-global self-distillation and soft cross-modal contrastive alignment strategies to enhance localization capabilities. Subsequently, we introduce an efficient cost aggregation module, comprising spatial and class aggregation mechanisms, to further advance dense prediction tasks with densely annotated datasets. Comprehensive evaluations on three tasks (phrase grounding, zero-shot semantic segmentation, and zero-shot classification) demonstrate DCXFM’s superior performance over other state-of-the-art medical image-text pretraining models. Remarkably, DCXFM exhibits powerful zero-shot capabilities across various datasets in phrase grounding and zero-shot semantic segmentation, underscoring its superior generalization in dense prediction tasks.
{"title":"Scaling Chest X-Ray Foundation Models From Mixed Supervisions for Dense Prediction","authors":"Fuying Wang;Lequan Yu","doi":"10.1109/TMI.2025.3589928","DOIUrl":"10.1109/TMI.2025.3589928","url":null,"abstract":"Foundation models have significantly revolutionized the field of chest X-ray diagnosis with their ability to transfer across various diseases and tasks. However, previous works have predominantly utilized self-supervised learning from medical image-text pairs, which falls short in dense medical prediction tasks due to their sole reliance on such coarse pair supervision, thereby limiting their applicability to detailed diagnostics. In this paper, we introduce a Dense Chest X-ray Foundation Model (DCXFM), which utilizes mixed supervision types (i.e., text, label, and segmentation masks) to significantly enhance the scalability of foundation models across various medical tasks. Our model involves two training stages: we first employ a novel self-distilled multimodal pretraining paradigm to exploit text and label supervision, along with local-to-global self-distillation and soft cross-modal contrastive alignment strategies to enhance localization capabilities. Subsequently, we introduce an efficient cost aggregation module, comprising spatial and class aggregation mechanisms, to further advance dense prediction tasks with densely annotated datasets. Comprehensive evaluations on three tasks (phrase grounding, zero-shot semantic segmentation, and zero-shot classification) demonstrate DCXFM’s superior performance over other state-of-the-art medical image-text pretraining models. Remarkably, DCXFM exhibits powerful zero-shot capabilities across various datasets in phrase grounding and zero-shot semantic segmentation, underscoring its superior generalization in dense prediction tasks.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 12","pages":"5306-5318"},"PeriodicalIF":0.0,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144645754","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}
Pub Date : 2025-07-16DOI: 10.1109/TMI.2025.3589946
Yuanwei He;Dan Ruan
Tomography recovers internal volume from projection measurements. Formulated as inverse problems, classic computed tomography generally reconstructs attenuation property in a preset cartesian grid coordinate. While this is intuitive and convenient for digital display, such discretization leads to forward-backward projection inconsistency, and discrepancy between digital and effective resolution. We take a different perspective by considering the image volume as continuous and modelling forward projection as a hybrid continuous-to-discrete mapping from volume to detector elements, which we call “ray bundles”. The ray bundle can be regarded as an unconventional heterogenous coordinate. Projections are modeled as line integrations along ray bundles in the continuous volume space and approximated by numerical integration using customized sample points. This modeling approach is conveniently supported with an implicit neural representation approach. By representing the volume as a function mapping spatial coordinates to attenuation properties and leveraging ray bundle projection, this approach reflects transmission physics and eliminates the need for explicit interpolation, intersection calculations, or matrix inversions. A novel sampling strategy is further developed to adaptively distribute points along the ray bundles, emphasizing high gradient regions to allocate computational resources to heterogenous structures and details. We call this system T-ReX to indicate Transmission Ray bundles for X-ray geometry. We validate T-ReX through comprehensive experiments across three scenarios: simulated full-fan projections with primary signal only, half-fan setups with simulated scatter and noise, and an in-house dataset with realistic acquisition conditions. These results highlight the effectiveness of T-ReX in sparse view X-ray tomography.
{"title":"Ray-Bundle-Based X-Ray Representation and Reconstruction: An Alternative to Classic Tomography on Voxelized Volumes","authors":"Yuanwei He;Dan Ruan","doi":"10.1109/TMI.2025.3589946","DOIUrl":"10.1109/TMI.2025.3589946","url":null,"abstract":"Tomography recovers internal volume from projection measurements. Formulated as inverse problems, classic computed tomography generally reconstructs attenuation property in a preset cartesian grid coordinate. While this is intuitive and convenient for digital display, such discretization leads to forward-backward projection inconsistency, and discrepancy between digital and effective resolution. We take a different perspective by considering the image volume as continuous and modelling forward projection as a hybrid continuous-to-discrete mapping from volume to detector elements, which we call “ray bundles”. The ray bundle can be regarded as an unconventional heterogenous coordinate. Projections are modeled as line integrations along ray bundles in the continuous volume space and approximated by numerical integration using customized sample points. This modeling approach is conveniently supported with an implicit neural representation approach. By representing the volume as a function mapping spatial coordinates to attenuation properties and leveraging ray bundle projection, this approach reflects transmission physics and eliminates the need for explicit interpolation, intersection calculations, or matrix inversions. A novel sampling strategy is further developed to adaptively distribute points along the ray bundles, emphasizing high gradient regions to allocate computational resources to heterogenous structures and details. We call this system T-ReX to indicate Transmission Ray bundles for X-ray geometry. We validate T-ReX through comprehensive experiments across three scenarios: simulated full-fan projections with primary signal only, half-fan setups with simulated scatter and noise, and an in-house dataset with realistic acquisition conditions. These results highlight the effectiveness of T-ReX in sparse view X-ray tomography.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 12","pages":"5128-5141"},"PeriodicalIF":0.0,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144645752","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}
Pub Date : 2025-07-16DOI: 10.1109/TMI.2025.3589543
Shixuan Leslie Gu;Jason Ken Adhinarta;Mikhail Bessmeltsev;Jiancheng Yang;Yongjie Jessica Zhang;Wenjie Yin;Daniel Berger;Jeff W. Lichtman;Hanspeter Pfister;Donglai Wei
Accurate segmentation of anatomical substructures within 3D curvilinear structures in medical imaging remains challenging due to their complex geometry and the scarcity of diverse, large-scale datasets for algorithm development and evaluation. In this paper, we use dendritic spine segmentation as a case study and address these challenges by introducing a novel Frenet-Serret Frame-based Decomposition, which decomposes 3D curvilinear structures into a globally smooth continuous curve that captures the overall shape, and a cylindrical primitive that encodes local geometric properties. This approach leverages Frenet-Serret Frames and arc length parameterization to preserve essential geometric features while reducing representational complexity, facilitating data-efficient learning, improved segmentation accuracy, and generalization on 3D curvilinear structures. To rigorously evaluate our method, we introduce two datasets: CurviSeg, a synthetic dataset for 3D curvilinear structure segmentation that validates our method’s key properties, and DenSpineEM, a benchmark for dendritic spine segmentation, which comprises 4,476 manually annotated spines from 70 dendrites across three public electron microscopy datasets, covering multiple brain regions and species. Our experiments on DenSpineEM demonstrate exceptional cross-region and cross-species generalization: models trained on the mouse somatosensory cortex subset achieve 94.43% Dice, maintaining strong performance in zero-shot segmentation on both mouse visual cortex (95.61% Dice) and human frontal lobe (86.63% Dice) subsets. Moreover, we test the generalizability of our method on the IntrA dataset, where it achieves 77.08% Dice (5.29% higher than prior arts) on intracranial aneurysm segmentation from entire artery models. These findings demonstrate the potential of our approach for accurately analyzing complex curvilinear structures across diverse medical imaging fields. Our dataset, code, and models are available at https://github.com/VCG/FFD4DenSpineEM to support future research.
{"title":"Frenet–Serret Frame-Based Decomposition for Part Segmentation of 3-D Curvilinear Structures","authors":"Shixuan Leslie Gu;Jason Ken Adhinarta;Mikhail Bessmeltsev;Jiancheng Yang;Yongjie Jessica Zhang;Wenjie Yin;Daniel Berger;Jeff W. Lichtman;Hanspeter Pfister;Donglai Wei","doi":"10.1109/TMI.2025.3589543","DOIUrl":"10.1109/TMI.2025.3589543","url":null,"abstract":"Accurate segmentation of anatomical substructures within 3D curvilinear structures in medical imaging remains challenging due to their complex geometry and the scarcity of diverse, large-scale datasets for algorithm development and evaluation. In this paper, we use dendritic spine segmentation as a case study and address these challenges by introducing a novel Frenet-Serret Frame-based Decomposition, which decomposes 3D curvilinear structures into a globally smooth continuous curve that captures the overall shape, and a cylindrical primitive that encodes local geometric properties. This approach leverages Frenet-Serret Frames and arc length parameterization to preserve essential geometric features while reducing representational complexity, facilitating data-efficient learning, improved segmentation accuracy, and generalization on 3D curvilinear structures. To rigorously evaluate our method, we introduce two datasets: CurviSeg, a synthetic dataset for 3D curvilinear structure segmentation that validates our method’s key properties, and DenSpineEM, a benchmark for dendritic spine segmentation, which comprises 4,476 manually annotated spines from 70 dendrites across three public electron microscopy datasets, covering multiple brain regions and species. Our experiments on DenSpineEM demonstrate exceptional cross-region and cross-species generalization: models trained on the mouse somatosensory cortex subset achieve 94.43% Dice, maintaining strong performance in zero-shot segmentation on both mouse visual cortex (95.61% Dice) and human frontal lobe (86.63% Dice) subsets. Moreover, we test the generalizability of our method on the IntrA dataset, where it achieves 77.08% Dice (5.29% higher than prior arts) on intracranial aneurysm segmentation from entire artery models. These findings demonstrate the potential of our approach for accurately analyzing complex curvilinear structures across diverse medical imaging fields. Our dataset, code, and models are available at <uri>https://github.com/VCG/FFD4DenSpineEM</uri> to support future research.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 12","pages":"5319-5331"},"PeriodicalIF":0.0,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144645751","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}
Pub Date : 2025-07-15DOI: 10.1109/TMI.2025.3589399
Boxiang Yun;Shitian Zhao;Qingli Li;Alex Kot;Yan Wang
With the assistance of large language models, which offer universal medical prior knowledge via text prompts, state-of-the-art Universal Models (UM) have demonstrated considerable potential in the field of medical image segmentation. Semantically detailed text prompts, on the one hand, indicate comprehensive knowledge; on the other hand, they bring biases that may not be applicable to specific cases involving heterogeneous organs or rare cancers. To this end, we propose a Debiased Universal Model (DUM) to consider instance-level context information and remove knowledge biases in text prompts from the causal perspective. We are the first to discover and mitigate the bias introduced by universal knowledge. Specifically, we propose to extract organ-level text prompts via language models and instance-level context prompts from the visual features of each image. We aim to highlight more on factual instance-level information and mitigate organ-level’s knowledge bias. This process can be derived and theoretically supported by a causal graph, and instantiated by designing a standard UM (SUM) and a biased UM. The debiased output is finally obtained by subtracting the likelihood distribution output by biased UM from that of the SUM. Experiments on three large-scale multi-center external datasets and MSD internal tumor datasets show that our method enhances the model’s generalization ability in handling diverse medical scenarios and reducing the potential biases, even with an improvement of 4.16% compared with popular universal model on the AbdomenAtlas dataset, showing the strong generalizability. The code is publicly available at https://github.com/DeepMed-Lab-ECNU/DUM
在通过文本提示提供通用医学先验知识的大型语言模型的帮助下,最先进的通用模型(UM)在医学图像分割领域显示出相当大的潜力。语义详实的文本提示,一方面表明知识全面;另一方面,它们带来的偏见可能不适用于涉及异质器官或罕见癌症的特定病例。为此,我们提出了一个Debiased Universal Model (DUM)来考虑实例级上下文信息,并从因果关系的角度消除文本提示中的知识偏差。我们是第一个发现并减轻普遍知识带来的偏见的人。具体来说,我们建议通过语言模型和实例级上下文提示从每个图像的视觉特征中提取器官级文本提示。我们的目标是更多地强调事实的实例级信息,减轻器官级的知识偏差。这一过程可以由因果图推导和理论上支持,并通过设计一个标准UM (SUM)和一个有偏UM来实例化。通过从SUM的似然分布输出中减去有偏UM的似然分布输出,最终得到无偏输出。在三个大规模多中心外部数据集和MSD内部肿瘤数据集上的实验表明,我们的方法增强了模型处理多种医疗场景的泛化能力,减少了潜在的偏差,甚至比目前流行的通用模型在腹大图数据集上的泛化能力提高了4.16%,显示出较强的泛化能力。该代码可在https://github.com/DeepMed-Lab-ECNU/DUM上公开获得
{"title":"Debiasing Medical Knowledge for Prompting Universal Model in CT Image Segmentation","authors":"Boxiang Yun;Shitian Zhao;Qingli Li;Alex Kot;Yan Wang","doi":"10.1109/TMI.2025.3589399","DOIUrl":"10.1109/TMI.2025.3589399","url":null,"abstract":"With the assistance of large language models, which offer universal medical prior knowledge via text prompts, state-of-the-art Universal Models (UM) have demonstrated considerable potential in the field of medical image segmentation. Semantically detailed text prompts, on the one hand, indicate comprehensive knowledge; on the other hand, they bring biases that may not be applicable to specific cases involving heterogeneous organs or rare cancers. To this end, we propose a Debiased Universal Model (DUM) to consider instance-level context information and remove knowledge biases in text prompts from the causal perspective. We are the first to discover and mitigate the bias introduced by universal knowledge. Specifically, we propose to extract organ-level text prompts via language models and instance-level context prompts from the visual features of each image. We aim to highlight more on factual instance-level information and mitigate organ-level’s knowledge bias. This process can be derived and theoretically supported by a causal graph, and instantiated by designing a standard UM (SUM) and a biased UM. The debiased output is finally obtained by subtracting the likelihood distribution output by biased UM from that of the SUM. Experiments on three large-scale multi-center external datasets and MSD internal tumor datasets show that our method enhances the model’s generalization ability in handling diverse medical scenarios and reducing the potential biases, even with an improvement of 4.16% compared with popular universal model on the AbdomenAtlas dataset, showing the strong generalizability. The code is publicly available at <uri>https://github.com/DeepMed-Lab-ECNU/DUM</uri>","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 12","pages":"5142-5154"},"PeriodicalIF":0.0,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144639749","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}
Pub Date : 2025-07-15DOI: 10.1109/TMI.2025.3587733
Zheang Huai;Hui Tang;Yi Li;Zhuangzhuang Chen;Xiaomeng Li
Source-free domain adaptation (SFDA) for segmentation aims at adapting a model trained in the source domain to perform well in the target domain with only the source model and unlabeled target data. Inspired by the recent success of Segment Anything Model (SAM) which exhibits the generality of segmenting images of various modalities and in different domains given human-annotated prompts like bounding boxes or points, we for the first time explore the potentials of Segment Anything Model for SFDA via automatedly finding an accurate bounding box prompt. We find that the bounding boxes directly generated with existing SFDA approaches are defective due to the domain gap. To tackle this issue, we propose a novel Dual Feature Guided (DFG) auto-prompting approach to search for the box prompt. Specifically, the source model is first trained in a feature aggregation phase, which not only preliminarily adapts the source model to the target domain but also builds a feature distribution well-prepared for box prompt search. In the second phase, based on two feature distribution observations, we gradually expand the box prompt with the guidance of the target model feature and the SAM feature to handle the class-wise clustered target features and the class-wise dispersed target features, respectively. To remove the potentially enlarged false positive regions caused by the over-confident prediction of the target model, the refined pseudo-labels produced by SAM are further postprocessed based on connectivity analysis. Experiments on 3D and 2D datasets indicate that our approach yields superior performance compared to conventional methods. Code is available at https://github.com/xmed-lab/DFG.
无源域自适应(source -free domain adaptation, SFDA)分割的目的是使在源域中训练好的模型在只有源模型和未标记的目标数据的情况下在目标域中表现良好。受到最近成功的Segment Anything Model (SAM)的启发,我们首次通过自动找到准确的边界框提示来探索Segment Anything Model在SFDA中的潜力。SAM展示了在给定的人类注释提示(如边界框或点)下,对各种模式和不同领域的图像进行分割的通用性。我们发现用现有的SFDA方法直接生成的边界盒由于域间隙存在缺陷。为了解决这个问题,我们提出了一种新的双特征引导(DFG)自动提示方法来搜索框提示符。具体来说,首先在特征聚合阶段对源模型进行训练,不仅使源模型初步适应目标域,而且构建了一个为框提示搜索做好准备的特征分布。第二阶段,在两次特征分布观测的基础上,在目标模型特征和SAM特征的指导下,逐步扩展框提示,分别处理类明智的聚类目标特征和类明智的分散目标特征。为了去除由于对目标模型的过度自信预测而可能增大的假阳性区域,对由SAM生成的精细伪标签进行基于连通性分析的进一步后处理。在3D和2D数据集上的实验表明,与传统方法相比,我们的方法具有更好的性能。代码可从https://github.com/xmed-lab/DFG获得。
{"title":"Leveraging Segment Anything Model for Source-Free Domain Adaptation via Dual Feature Guided Auto-Prompting","authors":"Zheang Huai;Hui Tang;Yi Li;Zhuangzhuang Chen;Xiaomeng Li","doi":"10.1109/TMI.2025.3587733","DOIUrl":"10.1109/TMI.2025.3587733","url":null,"abstract":"Source-free domain adaptation (SFDA) for segmentation aims at adapting a model trained in the source domain to perform well in the target domain with only the source model and unlabeled target data. Inspired by the recent success of Segment Anything Model (SAM) which exhibits the generality of segmenting images of various modalities and in different domains given human-annotated prompts like bounding boxes or points, we for the first time explore the potentials of Segment Anything Model for SFDA via automatedly finding an accurate bounding box prompt. We find that the bounding boxes directly generated with existing SFDA approaches are defective due to the domain gap. To tackle this issue, we propose a novel Dual Feature Guided (DFG) auto-prompting approach to search for the box prompt. Specifically, the source model is first trained in a feature aggregation phase, which not only preliminarily adapts the source model to the target domain but also builds a feature distribution well-prepared for box prompt search. In the second phase, based on two feature distribution observations, we gradually expand the box prompt with the guidance of the target model feature and the SAM feature to handle the class-wise clustered target features and the class-wise dispersed target features, respectively. To remove the potentially enlarged false positive regions caused by the over-confident prediction of the target model, the refined pseudo-labels produced by SAM are further postprocessed based on connectivity analysis. Experiments on 3D and 2D datasets indicate that our approach yields superior performance compared to conventional methods. Code is available at <uri>https://github.com/xmed-lab/DFG</uri>.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 12","pages":"5077-5088"},"PeriodicalIF":0.0,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144639870","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}
Colorectal cancer (CRC) is a significant global health concern, and early detection through screening plays a critical role in reducing mortality. While deep learning models have shown promise in improving polyp detection, classification, and segmentation, their generalization across diverse clinical environments, particularly with out-of-distribution (OOD) data, remains a challenge. Multi-center datasets like PolypGen have been developed to address these issues, but their collection is costly and time-consuming. Traditional data augmentation techniques provide limited variability, failing to capture the complexity of medical images. Diffusion models have emerged as a promising solution for generating synthetic polyp images, but the image generation process in current models mainly relies on segmentation masks as the condition, limiting their ability to capture the full clinical context. To overcome these limitations, we propose a Progressive Spectrum Diffusion Model (PSDM) that integrates diverse clinical annotations–such as segmentation masks, bounding boxes, and colonoscopy reports–by transforming them into compositional prompts. These prompts are organized into coarse and fine components, allowing the model to capture both broad spatial structures and fine details, generating clinically accurate synthetic images. By augmenting training data with PSDM-generated samples, our model significantly improves polyp detection, classification, and segmentation. For instance, on the PolypGen dataset, PSDM increases the F1 score by 2.12% and the mean average precision by 3.09%, demonstrating superior performance in OOD scenarios and enhanced generalization.
{"title":"Robust Polyp Detection and Diagnosis Through Compositional Prompt-Guided Diffusion Models","authors":"Jia Yu;Yan Zhu;Peiyao Fu;Tianyi Chen;Junbo Huang;Quanlin Li;Pinghong Zhou;Zhihua Wang;Fei Wu;Shuo Wang;Xian Yang","doi":"10.1109/TMI.2025.3589456","DOIUrl":"10.1109/TMI.2025.3589456","url":null,"abstract":"Colorectal cancer (CRC) is a significant global health concern, and early detection through screening plays a critical role in reducing mortality. While deep learning models have shown promise in improving polyp detection, classification, and segmentation, their generalization across diverse clinical environments, particularly with out-of-distribution (OOD) data, remains a challenge. Multi-center datasets like PolypGen have been developed to address these issues, but their collection is costly and time-consuming. Traditional data augmentation techniques provide limited variability, failing to capture the complexity of medical images. Diffusion models have emerged as a promising solution for generating synthetic polyp images, but the image generation process in current models mainly relies on segmentation masks as the condition, limiting their ability to capture the full clinical context. To overcome these limitations, we propose a Progressive Spectrum Diffusion Model (PSDM) that integrates diverse clinical annotations–such as segmentation masks, bounding boxes, and colonoscopy reports–by transforming them into compositional prompts. These prompts are organized into coarse and fine components, allowing the model to capture both broad spatial structures and fine details, generating clinically accurate synthetic images. By augmenting training data with PSDM-generated samples, our model significantly improves polyp detection, classification, and segmentation. For instance, on the PolypGen dataset, PSDM increases the F1 score by 2.12% and the mean average precision by 3.09%, demonstrating superior performance in OOD scenarios and enhanced generalization.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 12","pages":"5245-5257"},"PeriodicalIF":0.0,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11080481","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144639753","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}
Lumbar disc degeneration, a progressive structural wear and tear of lumbar intervertebral disc, is regarded as an essential role on low back pain, a significant global health concern. Automated lumbar spine geometry reconstruction from MR images will enable fast measurement of medical parameters to evaluate the lumbar status, in order to determine a suitable treatment. Existing image segmentation-based techniques often generate erroneous segments or unstructured point clouds, unsuitable for medical parameter measurement. In this work, we present UNet-DeformSA and TransDeformer: novel attention-based deep neural networks that reconstruct the geometry of the lumbar spine with high spatial accuracy and mesh correspondence across patients, and we also present a variant of TransDeformer for error estimation. Specially, we devise new attention modules with a new attention formula, which integrate tokenized image features and tokenized shape features to predict the displacements of the points on a shape template. The deformed template reveals the lumbar spine geometry in an image. Experiment results show that our networks generate artifact-free geometry outputs, and the variant of TransDeformer can predict the errors of a reconstructed geometry. Our code is available at https://github.com/linchenq/TransDeformer-Mesh.
{"title":"Attention-Based Shape-Deformation Networks for Artifact-Free Geometry Reconstruction of Lumbar Spine From MR Images","authors":"Linchen Qian;Jiasong Chen;Linhai Ma;Timur Urakov;Weiyong Gu;Liang Liang","doi":"10.1109/TMI.2025.3588831","DOIUrl":"10.1109/TMI.2025.3588831","url":null,"abstract":"Lumbar disc degeneration, a progressive structural wear and tear of lumbar intervertebral disc, is regarded as an essential role on low back pain, a significant global health concern. Automated lumbar spine geometry reconstruction from MR images will enable fast measurement of medical parameters to evaluate the lumbar status, in order to determine a suitable treatment. Existing image segmentation-based techniques often generate erroneous segments or unstructured point clouds, unsuitable for medical parameter measurement. In this work, we present UNet-DeformSA and TransDeformer: novel attention-based deep neural networks that reconstruct the geometry of the lumbar spine with high spatial accuracy and mesh correspondence across patients, and we also present a variant of TransDeformer for error estimation. Specially, we devise new attention modules with a new attention formula, which integrate tokenized image features and tokenized shape features to predict the displacements of the points on a shape template. The deformed template reveals the lumbar spine geometry in an image. Experiment results show that our networks generate artifact-free geometry outputs, and the variant of TransDeformer can predict the errors of a reconstructed geometry. Our code is available at <uri>https://github.com/linchenq/TransDeformer-Mesh</uri>.","PeriodicalId":94033,"journal":{"name":"IEEE transactions on medical imaging","volume":"44 12","pages":"5258-5277"},"PeriodicalIF":0.0,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144639750","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}