Medical image analysis最新文献_第3页

Detecting low-amplitude biomarker activations via decomposition of complex-valued fMRI data with collaborative phase and magnitude sparsity 通过分解具有协同相位和量级稀疏性的复杂值fMRI数据来检测低幅度生物标志物激活。

IF 11.8 1区医学

Medical image analysis Pub Date : 2025-09-07 DOI: 10.1016/j.media.2025.103803

Jia-Yang Song , Qiu-Hua Lin , Chi Zhou , Yi-Ran Wang , Yu-Ping Wang , Vince D. Calhoun

{"title":"Detecting low-amplitude biomarker activations via decomposition of complex-valued fMRI data with collaborative phase and magnitude sparsity","authors":"Jia-Yang Song , Qiu-Hua Lin , Chi Zhou , Yi-Ran Wang , Yu-Ping Wang , Vince D. Calhoun","doi":"10.1016/j.media.2025.103803","DOIUrl":"10.1016/j.media.2025.103803","url":null,"abstract":"<div><div>Sparse decomposition of complex-valued functional magnetic resonance imaging (fMRI) data is promising in finding qualified biomarkers for brain disorders such as schizophrenia, by simultaneously using intrinsic spatial sparsity and full functional information of the brain. However, previous methods may miss disease-related low-amplitude activations, since it is challenging to determine if a low-amplitude voxel is signal or noise during the iterative update process based solely on magnitude or phase sparsity. To this end, we propose a novel sparse decomposition model with collaborative phase and magnitude sparsity constraints at the voxel level. Specifically, we impose a sparsity constraint on the product of the magnitude and phase of a voxel above a pre-defined phase threshold. The low-amplitude activations with larger phase changes can survive the update process, despite temporarily violating the small-phase-change characteristic of signal voxels. Moreover, we eliminate phase ambiguity during iterations by proving no additional phase change is introduced by the update rules and by initializing the dictionary matrix atoms using the observed time series with fixed phase angles. We evaluate the proposed method using complex-valued simulated data and experimental resting-state fMRI data from schizophrenia patients and healthy controls. Compared with three state-of-the-art algorithms, the proposed method retains more low-amplitude activations in biomarker regions such as the anterior cingulate cortex and yields sensitive phase maps to disease-related spatial changes. This provides a new tool to estimate an informative fMRI biomarker of mental disorders.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103803"},"PeriodicalIF":11.8,"publicationDate":"2025-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145149769","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Trimming-then-augmentation: Towards robust depth and odometry estimation for endoscopic images 修剪-然后增强：对内窥镜图像的鲁棒深度和里程估计

IF 11.8 1区医学

Medical image analysis Pub Date : 2025-09-03 DOI: 10.1016/j.media.2025.103736

Junyang Wu , Yun Gu , Guang-Zhong Yang

{"title":"Trimming-then-augmentation: Towards robust depth and odometry estimation for endoscopic images","authors":"Junyang Wu , Yun Gu , Guang-Zhong Yang","doi":"10.1016/j.media.2025.103736","DOIUrl":"10.1016/j.media.2025.103736","url":null,"abstract":"<div><div>Depth and odometry estimation for endoscopic imaging is an essential task for robot assisted endoluminal intervention. Due to the difficulty of obtaining sufficient <em>in vivo</em> ground truth data, unsupervised learning is preferred in practical settings. Existing methods, however, are hampered by imaging artifacts and the paucity of unique anatomical markers, coupled with tissue motion and specular reflections, leading to the poor accuracy and generalizability. In this work, a trimming-then-augmentation framework is proposed. It uses a “mask-then-recover” training strategy to firstly mask out the artifact regions and then reconstruct the depth and pose information based on the global perception of a convolutional network. Subsequently, an augmentation module is used to provide stable correspondence between endoscopic image pairs. A task-specific loss function guides the augmentation module to adaptively establish stable feature pairs for improving the overall accuracy of subsequent 3D structural reconstruction. Detailed validation has been performed with results showing that the proposed method can significantly improve the accuracy of existing state-of-the-art unsupervised methods, demonstrating the effectiveness of the method and its resilience to image artifacts, in addition to its stability when applied to <em>in vivo</em> settings.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103736"},"PeriodicalIF":11.8,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Reinforced physiology-informed learning for image completion from partial-frame dynamic PET imaging 从部分帧动态PET成像增强生理信息学习图像补全

IF 11.8 1区医学

Medical image analysis Pub Date : 2025-09-03 DOI: 10.1016/j.media.2025.103767

Hengjia Ran , Jianan Cui , Xuhui Feng , Yubo Ye , Yufei Jin , Yunmei Chen , Bo Zhao , Rui Hu , Min Guo , Xinhui Su , Huafeng Liu

{"title":"Reinforced physiology-informed learning for image completion from partial-frame dynamic PET imaging","authors":"Hengjia Ran , Jianan Cui , Xuhui Feng , Yubo Ye , Yufei Jin , Yunmei Chen , Bo Zhao , Rui Hu , Min Guo , Xinhui Su , Huafeng Liu","doi":"10.1016/j.media.2025.103767","DOIUrl":"10.1016/j.media.2025.103767","url":null,"abstract":"<div><div>Dynamic positron emission tomography(PET) imaging using <span><math><msup><mrow></mrow><mrow><mn>18</mn></mrow></msup></math></span>F-FDG typically requires over an hour to acquire a complete time series of images. Therefore, reducing dynamic PET scan time is crucial for minimizing errors caused by patient movement and increasing the throughput of the imaging equipment. However, shortening the scanning time will lead to the loss of images in some frames, affecting the accuracy of PET parameter estimation. In this paper, we proposed a method that combined physiology-informed learning with time-implicit neural representations for kinetic modeling and missing-frame dynamic PET image completion. Based on the two-tissue compartment model, three types of constraint terms were constructed for network training, including data terms, boundary terms, and reinforced physiology residual terms. The method works effectively without the need for specific training datasets, making it feasible even with limited data. Three commonly used scanning schemes were defined to verify the feasibility of the proposed method and the performance was evaluated based on simulation data and real rat data. The best-performing scheme was selected for detailed analysis of PET images and parameter maps on datasets of four human organs obtained from Biograph Vision Quadra. Our method outperforms traditional nonlinear least squares (NLLS) fitting in both reconstruction quality and computational efficiency. The metrics calculated from different organs, such as the brain (SSIM <span><math><mo>></mo></math></span> 0.98) and the thorax (PSNR <span><math><mo>></mo></math></span> 40), show that the proposed network can achieve promising performance.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103767"},"PeriodicalIF":11.8,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145047698","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Spatio-temporal registration of multi-perspective 3D echocardiography for improved strain estimation 多视角三维超声心动图的时空配准改进应变估计

IF 11.8 1区医学

Medical image analysis Pub Date : 2025-09-02 DOI: 10.1016/j.media.2025.103791

M. Sjoerdsma , S. Bouwmeester , F. van Heesch , P. Houthuizen , R.G.P. Lopata

{"title":"Spatio-temporal registration of multi-perspective 3D echocardiography for improved strain estimation","authors":"M. Sjoerdsma , S. Bouwmeester , F. van Heesch , P. Houthuizen , R.G.P. Lopata","doi":"10.1016/j.media.2025.103791","DOIUrl":"10.1016/j.media.2025.103791","url":null,"abstract":"<div><div>For heart diagnostics, ultrasound is generally the modality of choice due to its high temporal and spatial resolution, availability, and patient safety. Although 3D echocardiography captures the complex shape and motion of the heart with more precision than 2D, it suffers to a greater extent from poor resolution, noise, and limited field-of-view. Multi-perspective echocardiography has proven to significantly enhance both image quality and field-of-view. The greatest improvements occur when combining acquisitions from widely differing insonification angles, but this process is challenging because of substantial local structural and brightness variations and ultrasound’s anisotropic nature. To handle these inconsistencies, a novel temporal and spatial registration algorithm designed is proposed. Temporal registration is achieved using low-frequency cardiac wall features and motion extracted via singular value decomposition of a spatio-temporal Casorati matrix, while spatial registration is performed using phase-only correlation of low-frequency data. The acquisitions are seamlessly fused using a 3D, oriented, wavelet transform including a near-field clutter algorithm. <em>In vitro</em> and <em>in vivo</em> testing highlights the benefits of this approach. Temporal alignment, validated against electrocardiograms, is precise, with an average error of just 2 ± 10 ms. Furthermore, our method outperforms a six-degree-of-freedom encoder-based probe tracker, reducing spatial registration error to 5 ± 3 mm from 19 ± 10 mm. The resulting longitudinal and radial strain measurements closely align with those obtained by tagged magnetic resonance imaging, demonstrating the accuracy and feasibility of this technique.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103791"},"PeriodicalIF":11.8,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145047578","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Developing a knowledge-guided federated graph attention learning network with a diffusion module to diagnose Alzheimer’s disease 开发一个带扩散模块的知识引导联邦图注意力学习网络用于阿尔茨海默病诊断。

IF 11.8 1区医学

Medical image analysis Pub Date : 2025-09-02 DOI: 10.1016/j.media.2025.103794

Xuegang Song , Kaixiang Shu , Peng Yang , Cheng Zhao , Feng Zhou , Alejandro F Frangi , Jiuwen Cao , Xiaohua Xiao , Shuqiang Wang , Tianfu Wang , Baiying Lei , Alzheimer’s Disease Neuroimaging Initiative

{"title":"Developing a knowledge-guided federated graph attention learning network with a diffusion module to diagnose Alzheimer’s disease","authors":"Xuegang Song , Kaixiang Shu , Peng Yang , Cheng Zhao , Feng Zhou , Alejandro F Frangi , Jiuwen Cao , Xiaohua Xiao , Shuqiang Wang , Tianfu Wang , Baiying Lei , Alzheimer’s Disease Neuroimaging Initiative","doi":"10.1016/j.media.2025.103794","DOIUrl":"10.1016/j.media.2025.103794","url":null,"abstract":"<div><div>In studies of Alzheimer’s disease (AD), limited sample size considerably hampers the performance of intelligent diagnostic systems. Using multi-site data increases sample size but raises concerns regarding data privacy and inter-site heterogeneity. To address these issues, we developed a knowledge-guided federated graph attention learning network with a diffusion module to facilitate AD diagnosis from multi-site data. We used multiple templates to extract regions-of-interest (ROI)-based volume features from structural magnetic resonance imaging (sMRI) data. These volume features were then combined with previously identified AD features from published studies (prior knowledge) to determine the discriminative features within the images. We then designed an attention-guided diffusion module to synthesize samples by prioritizing these key features. The diffusion module was trained within a federated learning framework, which ensured inter-site data privacy while limiting data heterogeneity. Finally, we designed a federated graph attention learning network as a classifier to capture AD-related deep features and improve the accuracy of diagnosing AD. The efficacy of our approach was validated using three AD datasets. Thus, the classifier developed in this study represents a promising tool for optimizing multi-site neuroimaging data to improving the accuracy of diagnosing AD in the clinic.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103794"},"PeriodicalIF":11.8,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145081237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

EndoChat: Grounded multimodal large language model for endoscopic surgery EndoChat：内窥镜手术的多模态大语言模型。

IF 11.8 1区医学

Medical image analysis Pub Date : 2025-08-31 DOI: 10.1016/j.media.2025.103789

Guankun Wang , Long Bai , Junyi Wang , Kun Yuan , Zhen Li , Tianxu Jiang , Xiting He , Jinlin Wu , Zhen Chen , Zhen Lei , Hongbin Liu , Jiazheng Wang , Fan Zhang , Nicolas Padoy , Nassir Navab , Hongliang Ren

{"title":"EndoChat: Grounded multimodal large language model for endoscopic surgery","authors":"Guankun Wang , Long Bai , Junyi Wang , Kun Yuan , Zhen Li , Tianxu Jiang , Xiting He , Jinlin Wu , Zhen Chen , Zhen Lei , Hongbin Liu , Jiazheng Wang , Fan Zhang , Nicolas Padoy , Nassir Navab , Hongliang Ren","doi":"10.1016/j.media.2025.103789","DOIUrl":"10.1016/j.media.2025.103789","url":null,"abstract":"<div><div>Recently, Multimodal Large Language Models (MLLMs) have demonstrated their immense potential in computer-aided diagnosis and decision-making. In the context of robotic-assisted surgery, MLLMs can serve as effective tools for surgical training and guidance. However, there is still a deficiency of MLLMs specialized for surgical scene understanding in endoscopic procedures. To this end, we present EndoChat, an MLLM tailored to address various dialogue paradigms and subtasks in understanding endoscopic procedures. To train our EndoChat, we construct the Surg-396K dataset through a novel pipeline that systematically extracts surgical information and generates structured annotations based on large-scale endoscopic surgery datasets. Furthermore, we introduce a multi-scale visual token interaction mechanism and a visual contrast-based reasoning mechanism to enhance the model’s representation learning and reasoning capabilities. Our model achieves state-of-the-art performance across five dialogue paradigms and seven surgical scene understanding tasks. Additionally, we conduct evaluations with professional surgeons, who provide positive feedback on the majority of conversation cases generated by EndoChat. Overall, these results demonstrate that EndoChat has the potential to advance training and automation in robotic-assisted surgery. Our dataset and model are publicly available at <span><span>https://github.com/gkw0010/EndoChat</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103789"},"PeriodicalIF":11.8,"publicationDate":"2025-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145033792","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Abn-BLIP: Abnormality-aligned Bootstrapping Language-Image Pre-training for pulmonary embolism diagnosis and report generation from CTPA Abn-BLIP：异常对齐引导语言图像预训练用于肺栓塞诊断和CTPA报告生成

IF 11.8 1区医学

Medical image analysis Pub Date : 2025-08-30 DOI: 10.1016/j.media.2025.103786

Zhusi Zhong , Yuli Wang , Lulu Bi , Zhuoqi Ma , Sun Ho Ahn , Christopher J. Mullin , Colin F. Greineder , Michael K. Atalay , Scott Collins , Grayson L. Baird , Cheng Ting Lin , J. Webster Stayman , Todd M. Kolb , Ihab Kamel , Harrison X. Bai , Zhicheng Jiao

{"title":"Abn-BLIP: Abnormality-aligned Bootstrapping Language-Image Pre-training for pulmonary embolism diagnosis and report generation from CTPA","authors":"Zhusi Zhong , Yuli Wang , Lulu Bi , Zhuoqi Ma , Sun Ho Ahn , Christopher J. Mullin , Colin F. Greineder , Michael K. Atalay , Scott Collins , Grayson L. Baird , Cheng Ting Lin , J. Webster Stayman , Todd M. Kolb , Ihab Kamel , Harrison X. Bai , Zhicheng Jiao","doi":"10.1016/j.media.2025.103786","DOIUrl":"10.1016/j.media.2025.103786","url":null,"abstract":"<div><div>Medical imaging plays a pivotal role in modern healthcare, with computed tomography pulmonary angiography (CTPA) being a critical tool for diagnosing pulmonary embolism and other thoracic conditions. However, the complexity of interpreting CTPA scans and generating accurate radiology reports remains a significant challenge. This paper introduces Abn-BLIP (Abnormality-aligned Bootstrapping Language-Image Pretraining), an advanced diagnosis model designed to align abnormal findings to generate the accuracy and comprehensiveness of radiology reports. By leveraging learnable queries and cross-modal attention mechanisms, our model demonstrates superior performance in detecting abnormalities, reducing missed findings, and generating structured reports compared to existing methods. Our experiments show that Abn-BLIP outperforms state-of-the-art medical vision-language models and 3D report generation methods in both accuracy and clinical relevance. These results highlight the potential of integrating multimodal learning strategies for improving radiology reporting. The source code is available at <span><span>https://github.com/zzs95/abn-blip</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103786"},"PeriodicalIF":11.8,"publicationDate":"2025-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145047691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Measuring and predicting where and when pathologists focus their visual attention while grading whole slide images of cancer 测量和预测病理学家在对整个癌症幻灯片图像进行分级时将视觉注意力集中在何处和何时

IF 11.8 1区医学

Medical image analysis Pub Date : 2025-08-28 DOI: 10.1016/j.media.2025.103752

Souradeep Chakraborty , Ruoyu Xue , Rajarsi Gupta , Oksana Yaskiv , Constantin Friedman , Natallia Sheuka , Dana Perez , Paul Friedman , Won-Tak Choi , Waqas Mahmud , Beatrice Knudsen , Gregory Zelinsky , Joel Saltz , Dimitris Samaras

{"title":"Measuring and predicting where and when pathologists focus their visual attention while grading whole slide images of cancer","authors":"Souradeep Chakraborty , Ruoyu Xue , Rajarsi Gupta , Oksana Yaskiv , Constantin Friedman , Natallia Sheuka , Dana Perez , Paul Friedman , Won-Tak Choi , Waqas Mahmud , Beatrice Knudsen , Gregory Zelinsky , Joel Saltz , Dimitris Samaras","doi":"10.1016/j.media.2025.103752","DOIUrl":"10.1016/j.media.2025.103752","url":null,"abstract":"<div><div>The ability to predict the attention of expert pathologists could lead to decision support systems for better pathology training. We developed methods to predict the spatio-temporal (“where” and “when”) movements of pathologists’ attention as they grade whole slide images (WSIs) of prostate cancer. We characterize a pathologist’s attention trajectory by their x, y, and m (magnification) movements of a viewport as they navigate WSIs using a digital microscope. This information was obtained from 43 pathologists across 123 WSIs, and we consider the task of predicting the pathologist attention scanpaths constructed from the viewport centers. We introduce a fixation extraction algorithm that simplifies an attention trajectory by extracting “fixations” in the pathologist’s viewing while preserving semantic information, and we use these pre-processed data to train and test a two-stage model to predict the dynamic (scanpath) allocation of attention during WSI reading via intermediate attention heatmap prediction. In the first stage, a transformer-based sub-network predicts the attention heatmaps (static attention) across different magnifications. In the second stage, we predict the attention scanpath by sequentially modeling the next fixation points in an autoregressive manner using a transformer-based approach, starting at the WSI center and leveraging multi-magnification feature representations from the first stage. Experimental results show that our scanpath prediction model outperforms chance and baseline models. Tools developed from this model could assist pathology trainees in learning to allocate their attention during WSI reading like an expert.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103752"},"PeriodicalIF":11.8,"publicationDate":"2025-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145047577","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

IterMask3D: Unsupervised anomaly detection and segmentation with test-time iterative mask refinement in 3D brain MRI IterMask3D：基于测试时间迭代掩模细化的三维脑MRI无监督异常检测与分割

IF 11.8 1区医学

Medical image analysis Pub Date : 2025-08-28 DOI: 10.1016/j.media.2025.103763

Ziyun Liang , Xiaoqing Guo , Wentian Xu , Yasin Ibrahim , Natalie Voets , Pieter M. Pretorius , Alzheimer’s Disease Neuroimaging Initiative, J. Alison Noble , Konstantinos Kamnitsas

{"title":"IterMask3D: Unsupervised anomaly detection and segmentation with test-time iterative mask refinement in 3D brain MRI","authors":"Ziyun Liang , Xiaoqing Guo , Wentian Xu , Yasin Ibrahim , Natalie Voets , Pieter M. Pretorius , Alzheimer’s Disease Neuroimaging Initiative, J. Alison Noble , Konstantinos Kamnitsas","doi":"10.1016/j.media.2025.103763","DOIUrl":"10.1016/j.media.2025.103763","url":null,"abstract":"<div><div>Unsupervised anomaly detection and segmentation methods train a model to learn the training distribution as ‘normal’. In the testing phase, they identify patterns that deviate from this normal distribution as ‘anomalies’. To learn the ‘normal’ distribution, prevailing methods corrupt the images and train a model to reconstruct them. During testing, the model attempts to reconstruct corrupted inputs based on the learned ‘normal’ distribution. Deviations from this distribution lead to high reconstruction errors, which indicate potential anomalies. However, corrupting an input image inevitably causes information loss even in normal regions, leading to suboptimal reconstruction and an increased risk of false positives. To alleviate this, we propose <span><math><mi>IterMask3D</mi></math></span>, an iterative spatial mask-refining strategy designed for 3D brain MRI. We iteratively spatially mask areas of the image as corruption and reconstruct them, then shrink the mask based on reconstruction error. This process iteratively unmasks ‘normal’ areas to the model, whose information further guides reconstruction of ‘normal’ patterns under the mask to be reconstructed accurately, reducing false positives. In addition, to achieve better reconstruction performance, we also propose using high-frequency image content as additional structural information to guide the reconstruction of the masked area. Extensive experiments on the detection of both synthetic and real-world imaging artifacts, as well as segmentation of various pathological lesions across multiple MRI sequences, consistently demonstrate the effectiveness of our proposed method. Code is available at <span><span>https://github.com/ZiyunLiang/IterMask3D</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103763"},"PeriodicalIF":11.8,"publicationDate":"2025-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145047697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

StaDis: Stability distance to detecting out-of-distribution data in computational pathology 计算病理学中检测非分布数据的稳定距离

IF 11.8 1区医学

Medical image analysis Pub Date : 2025-08-27 DOI: 10.1016/j.media.2025.103774

Di Zhang , Jiusong Ge , Jiashuai Liu , Chunbao Wang , Tieliang Gong , Zeyu Gao , Chen Li

{"title":"StaDis: Stability distance to detecting out-of-distribution data in computational pathology","authors":"Di Zhang , Jiusong Ge , Jiashuai Liu , Chunbao Wang , Tieliang Gong , Zeyu Gao , Chen Li","doi":"10.1016/j.media.2025.103774","DOIUrl":"10.1016/j.media.2025.103774","url":null,"abstract":"<div><div>Modern Computational pathology (CPath) models aim to alleviate the burden on pathologists. However, once deployed, these models may generate unreliable predictions when encountering data types not seen during training, potentially causing a trust crisis within the computational pathology community. Out-of-distribution (OOD) detection, acting as a safety measure before model deployment, demonstrates significant promise in ensuring the reliable use of models in real clinical application. However, most existing computational pathology models lack OOD detection mechanisms, and no OOD detection method is specifically designed for this field. In this paper, we propose a novel OOD detection approach called Stability Distance (StaDis), uniquely developed for CPath. StaDis measures the feature gap between an image and its perturbed counterpart. As a plug-and-play module, it requires no retraining and integrates seamlessly with any model. Additionally, for the first time, we explore OOD detection at the whole-slide image (WSI) level within the multiple instance learning (MIL) framework. Then, we design different pathological OOD detection benchmarks covering three real clinical scenarios: patch- and slide-level anomaly tissue detection, rare case mining, and frozen section (FS) detection. Finally, extensive comparative experiments are conducted on these pathological OOD benchmarks. In 38 experiments, our approach achieves SOTA performance in 23 cases and ranks second in 10 experiments. Especially, the AUROC results of StaDis with “Conch” as the backbone improve by 7.91% for patch-based anomaly tissue detection. Our code is available at <span><span>https://github.com/zdipath/StaDis</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103774"},"PeriodicalIF":11.8,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144917755","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0