Medical image analysis最新文献

{"title":"Uncertainty quantification for White Matter Hyperintensity segmentation detects silent failures and improves automated Fazekas quantification","authors":"Ben Philps ,&nbsp;Maria del C. Valdés Hernández ,&nbsp;Chen Qin ,&nbsp;Una Clancy ,&nbsp;Eleni Sakka ,&nbsp;Susana Muñoz Maniega ,&nbsp;Mark E. Bastin ,&nbsp;Angela C.C. Jochems ,&nbsp;Joanna M. Wardlaw ,&nbsp;Miguel O. Bernabeu ,&nbsp;Alzheimer’s Disease Neuroimaging Initiative (ADNI)","doi":"10.1016/j.media.2025.103697","DOIUrl":"10.1016/j.media.2025.103697","url":null,"abstract":"<div><div>White Matter Hyperintensities (WMH) are key neuroradiological markers of small vessel disease present in brain MRI. Assessment of WMH is important in research and clinics. However, WMH are challenging to segment due to their high variability in shape, location, size, poorly defined borders, and similar intensity profile to other pathologies (e.g stroke lesions) and artefacts (e.g head motion). In this work, we assess the utility and semantic properties of the most effective techniques for uncertainty quantification (UQ) in segmentation for the WMH segmentation task across multiple test-time data distributions. We find UQ techniques reduce ‘silent failure’ by identifying in UQ maps small WMH clusters in the deep white matter that are unsegmented by the model. A combination of Stochastic Segmentation Networks with Deep Ensembles also yields the highest Dice and lowest Absolute Volume Difference % (AVD) score and can highlight areas where there is ambiguity between WMH and stroke lesions. We further demonstrate the downstream utility of UQ, proposing a novel method for classification of the clinical Fazekas score using spatial features extracted from voxelwise WMH probability and UQ maps. We show that incorporating WMH uncertainty information improves Fazekas classification performance and calibration. Our model with (UQ and spatial WMH features)/(spatial WMH features)/(WMH volume only) achieves a balanced accuracy score of 0.74/0.67/0.62, and root brier score (<span><math><mi>↓</mi></math></span>) of 0.65/0.72/0.74 in the Deep WMH and balanced accuracy of 0.74/0.73/0.71 and root brier score of 0.64/0.66/0.68 in the Periventricular region. We further demonstrate that stochastic UQ techniques with high sample diversity can improve the detection of poor quality segmentations.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"105 ","pages":"Article 103697"},"PeriodicalIF":10.7,"publicationDate":"2025-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144622430","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}

{"title":"PathFL: Multi-alignment Federated Learning for pathology image segmentation","authors":"Yuan Zhang ,&nbsp;Feng Chen ,&nbsp;Yaolei Qi ,&nbsp;Guanyu Yang ,&nbsp;Huazhu Fu","doi":"10.1016/j.media.2025.103670","DOIUrl":"10.1016/j.media.2025.103670","url":null,"abstract":"<div><div>Pathology image segmentation across multiple centers encounters significant challenges due to diverse sources of heterogeneity including imaging modalities, organs, and scanning equipment, whose variability brings representation bias and impedes the development of generalizable segmentation models. In this paper, we propose PathFL, a novel multi-alignment Federated Learning framework for pathology image segmentation that addresses these challenges through three-level alignment strategies of image, feature, and model aggregation. Firstly, at the image level, a collaborative style enhancement module aligns and diversifies local data by facilitating style information exchange across clients. Secondly, at the feature level, an adaptive feature alignment module ensures implicit alignment in the representation space by infusing local features with global insights, promoting consistency across heterogeneous client features learning. Finally, at the model aggregation level, a stratified similarity aggregation strategy hierarchically aligns and aggregates models on the server, using layer-specific similarity to account for client discrepancies and enhance global generalization. Comprehensive evaluations on four sets of heterogeneous pathology image datasets, encompassing cross-source, cross-modality, cross-organ, and cross-scanner variations, validate the effectiveness of our PathFL in achieving better performance and robustness against data heterogeneity. The code is available at <span><span>https://github.com/yuanzhang7/PathFL</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"105 ","pages":"Article 103670"},"PeriodicalIF":10.7,"publicationDate":"2025-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144622432","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}

{"title":"Recursive variational autoencoders for 3D blood vessel generative modeling","authors":"Paula Feldman ,&nbsp;Miguel Fainstein ,&nbsp;Viviana Siless ,&nbsp;Claudio Delrieux ,&nbsp;Emmanuel Iarussi","doi":"10.1016/j.media.2025.103703","DOIUrl":"10.1016/j.media.2025.103703","url":null,"abstract":"<div><div>Anatomical trees play an important role in clinical diagnosis and treatment planning. Yet, accurately representing these structures poses significant challenges owing to their intricate and varied topology and geometry. Most existing methods to synthesize vasculature are rule based, and despite providing some degree of control and variation in the structures produced, they fail to capture the diversity and complexity of actual anatomical data. We developed a Recursive variational Neural Network (RvNN) that fully exploits the hierarchical organization of the vessel and learns a low-dimensional manifold encoding branch connectivity along with geometry features describing the target surface. After training, the RvNN latent space can be sampled to generate new vessel geometries. By leveraging the power of generative neural networks, we generate 3D models of blood vessels that are both accurate and diverse, which is crucial for medical and surgical training, hemodynamic simulations, and many other purposes. These results closely resemble real data, achieving high similarity in vessel radii, length, and tortuosity across various datasets, including those with aneurysms. To the best of our knowledge, this work is the first to utilize this technique for synthesizing blood vessels.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"105 ","pages":"Article 103703"},"PeriodicalIF":10.7,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144622434","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}

{"title":"Learning to segment anatomy and lesions from disparately labeled sources in brain MRI","authors":"Meva Himmetoglu, Ilja Ciernik, Ender Konukoglu","doi":"10.1016/j.media.2025.103705","DOIUrl":"https://doi.org/10.1016/j.media.2025.103705","url":null,"abstract":"Segmenting healthy tissue structures alongside lesions in brain Magnetic Resonance Images (MRI) remains a challenge for today’s algorithms due to lesion-caused disruption of the anatomy and lack of jointly labeled training datasets, where both healthy tissues and lesions are labeled on the same images. In this paper, we propose a method that is robust to lesion-caused disruptions and can be trained from disparately labeled training sets, i.e., without requiring jointly labeled samples, to automatically segment both. In contrast to prior work, we decouple healthy tissue and lesion segmentation in two paths to leverage multi-sequence acquisitions and merge information with an attention mechanism. During inference, an image-specific adaptation reduces adverse influences of lesion regions on healthy tissue predictions. During training, the adaptation is taken into account through meta-learning and co-training is used to learn from disparately labeled training images. Our model shows an improved performance on several anatomical structures and lesions on a publicly available brain glioblastoma dataset compared to the state-of-the-art segmentation methods.","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"109 1","pages":""},"PeriodicalIF":10.9,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144622436","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}

{"title":"Noise-inspired diffusion model for generalizable low-dose CT reconstruction","authors":"Qi Gao ,&nbsp;Zhihao Chen ,&nbsp;Dong Zeng ,&nbsp;Junping Zhang ,&nbsp;Jianhua Ma ,&nbsp;Hongming Shan","doi":"10.1016/j.media.2025.103710","DOIUrl":"10.1016/j.media.2025.103710","url":null,"abstract":"<div><div>The generalization of deep learning-based low-dose computed tomography (CT) reconstruction models to doses unseen in the training data is important and remains challenging. Previous efforts heavily rely on paired data to improve the generalization performance and robustness through collecting either diverse CT data for re-training or a few test data for fine-tuning. Recently, diffusion models have shown promising and generalizable performance in low-dose CT (LDCT) reconstruction, however, they may produce unrealistic structures due to the CT image noise deviating from Gaussian distribution and imprecise prior information from the guidance of noisy LDCT images. In this paper, we propose a <u>n</u>ois<u>e</u>-inspir<u>e</u>d <u>d</u>iffusion model for generalizable LDCT reconstruction, termed NEED, which tailors diffusion models for noise characteristics of each domain. First, we propose a novel shifted Poisson diffusion model to denoise projection data, which aligns the diffusion process with the noise model in pre-log LDCT projections. Second, we devise a doubly guided diffusion model to refine reconstructed images, which leverages LDCT images and initial reconstructions to more accurately locate prior information and enhance reconstruction fidelity. By cascading these two diffusion models for dual-domain reconstruction, our NEED requires only normal-dose data for training and can be effectively extended to various unseen dose levels during testing via a time step matching strategy. Extensive qualitative, quantitative, and segmentation-based evaluations on two datasets demonstrate that our NEED consistently outperforms state-of-the-art methods in reconstruction and generalization performance. Source code is made available at <span><span>https://github.com/qgao21/NEED</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"105 ","pages":"Article 103710"},"PeriodicalIF":10.7,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144605086","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}

{"title":"CLASS-M: Adaptive stain separation-based contrastive learning with pseudo-labeling for histopathological image classification","authors":"Bodong Zhang ,&nbsp;Hamid Manoochehri ,&nbsp;Man Minh Ho ,&nbsp;Fahimeh Fooladgar ,&nbsp;Yosep Chong ,&nbsp;Beatrice S. Knudsen ,&nbsp;Deepika Sirohi ,&nbsp;Tolga Tasdizen","doi":"10.1016/j.media.2025.103711","DOIUrl":"10.1016/j.media.2025.103711","url":null,"abstract":"<div><div>Histopathological image classification is an important task in medical image analysis. Recent approaches generally rely on weakly supervised learning due to the ease of acquiring case-level labels from pathology reports. However, patch-level classification is preferable in applications where only a limited number of cases are available or when local prediction accuracy is critical. On the other hand, acquiring extensive datasets with localized labels for training is not feasible. In this paper, we propose a semi-supervised patch-level histopathological image classification model, named CLASS-M, that does not require extensively labeled datasets. CLASS-M is formed by two main parts: a contrastive learning module that uses separated Hematoxylin images and Eosin images generated through an adaptive stain separation process, and a module with pseudo-labels using MixUp. We compare our model with other state-of-the-art models on two clear cell renal cell carcinoma datasets. We demonstrate that our CLASS-M model has the best performance on both datasets. Our code is available at <span><span>github.com/BzhangURU/Paper_CLASS-M/tree/main</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"105 ","pages":"Article 103711"},"PeriodicalIF":10.7,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144622435","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}

{"title":"Knowledge-driven interpretative conditional diffusion model for contrast-free myocardial infarction enhancement synthesis","authors":"Ronghui Qi ,&nbsp;Min Tao ,&nbsp;Chenchu Xu ,&nbsp;Xiaohu Li ,&nbsp;Siyuan Pan ,&nbsp;Jie Chen ,&nbsp;Shuo Li","doi":"10.1016/j.media.2025.103701","DOIUrl":"10.1016/j.media.2025.103701","url":null,"abstract":"<div><div>Synthesis of myocardial infarction enhancement (MIE) images without contrast agents (CAs) has shown great potential to advance myocardial infarction (MI) diagnosis and treatment. It provides results comparable to late gadolinium enhancement (LGE) images, thereby reducing the risks associated with CAs and streamlining clinical workflows. The existing knowledge-and-data-driven approach has made progress in addressing the complex challenges of synthesizing MIE images (i.e., invisible myocardial scars and high inter-individual variability) but still has limitations in the interpretability of kinematic inference, morphological knowledge integration, and kinematic-morphological fusion, thereby reducing the transparency and reliability of the model and causing information loss during synthesis. In this paper, we proposed a knowledge-driven interpretative conditional diffusion model (K-ICDM), which learns kinematic and morphological information from non-enhanced cardiac MR images (CINE sequence and T1 sequence) guided by cardiac knowledge, enabling the synthesis of MIE images. Importantly, our K-ICDM introduces three key innovations that address these limitations, thereby providing interpretability and improving synthesis quality. (1) A novel cardiac causal intervention that generates counterfactual strain to intervene in the inference process from motion maps to abnormal myocardial information, thereby establishing an explicit relationship and providing the clear causal interpretability. (2) A knowledge-driven cognitive combination strategy that utilizes cardiac signal topology knowledge to analyze T1 signal variations, enabling the model to understand how to learn morphological features, thus providing interpretability for morphology capture. (3) An information-specific adaptive fusion strategy that integrates kinematic and morphological information into the conditioning input of the diffusion model based on their specific contributions and adaptively learns their interactions, thereby preserving more detailed information. Experiments on a broad MI dataset with 315 patients show that our K-ICDM achieves state-of-the-art performance in contrast-free MIE image synthesis, improving structural similarity index measure (SSIM) by at least 2.1% over recent methods. These results demonstrate that our method effectively overcomes the limitations of existing methods in capturing the complex relationship between myocardial motion and scar distribution and integrating of static and dynamic sequences, thus enabling the accurate synthesis of subtle scar boundaries.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"105 ","pages":"Article 103701"},"PeriodicalIF":10.7,"publicationDate":"2025-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144587716","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}

{"title":"Vision-skeleton dual-modality framework for generalizable assessment of Parkinson’s disease gait","authors":"Weiping Liu ,&nbsp;Xiaozhen Lin ,&nbsp;Xinghong Chen ,&nbsp;Yifang Liu ,&nbsp;Zengxin Zhong ,&nbsp;Rong Chen ,&nbsp;Guannan Chen ,&nbsp;Yu Lin","doi":"10.1016/j.media.2025.103727","DOIUrl":"10.1016/j.media.2025.103727","url":null,"abstract":"<div><div>Gait abnormalities in Parkinson’s disease (PD) can reflect the extent of dysfunction, and making their assessment crucial for the diagnosis and treatment of PD. Current video-based methods of PD gait assessment are limited to only focusing on skeleton motion information and are confined to evaluations from a single perspective. To overcome these limitations, we propose a novel vision-skeleton dual-modality framework, which integrates keypoints vision features with skeleton motion information to enable a more accurate and comprehensive assessment of PD gait. We firstly introduce the Keypoints Vision Transformer, a novel architecture designed to extract vision features of human keypoints. This model encompasses both the spatial locations and connectivity relationships of human keypoints. Subsequently, through the proposed temporal fusion encoder, we integrate the extracted skeleton motion with keypoints vision features to enhance the extraction of temporal motion features. In a video dataset of 241 PD participants recorded from the front, our proposed framework achieves an assessment accuracy of 78.05%, which demonstrates superior performance compared to other methods. To enhance the interpretability of our method, we also conduct a feature visualization analysis of the proposed dual-modality framework, which reveal the mechanisms of different body parts and dual-modality branch in PD gait assessment. Additionally, when applied to another video dataset recorded from a more general perspective, our method still achieves a commendable accuracy of 73.07%. This achievement demonstrates the robust generalization capability of the proposed model in PD gait assessment from cross-view, which offers a novel approach for realizing unrestricted PD gait assessment in home monitoring. The latest version of the code is available at <span><span>https://github.com/FJNU-LWP/PD-gait-VSDF</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"105 ","pages":"Article 103727"},"PeriodicalIF":10.7,"publicationDate":"2025-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144614076","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}

{"title":"Hierarchical Vision Transformers for prostate biopsy grading: Towards bridging the generalization gap","authors":"Clément Grisi ,&nbsp;Kimmo Kartasalo ,&nbsp;Martin Eklund ,&nbsp;Lars Egevad ,&nbsp;Jeroen van der Laak ,&nbsp;Geert Litjens","doi":"10.1016/j.media.2025.103663","DOIUrl":"10.1016/j.media.2025.103663","url":null,"abstract":"<div><div>Practical deployment of Vision Transformers in computational pathology has largely been constrained by the sheer size of whole-slide images. Transformers faced a similar limitation when applied to long documents, and Hierarchical Transformers were introduced to circumvent it. This work explores the capabilities of Hierarchical Vision Transformers for prostate cancer grading in WSIs and presents a novel technique to combine attention scores smartly across hierarchical transformers. Our best-performing model matches state-of-the-art algorithms with a 0.916 quadratic kappa on the Prostate cANcer graDe Assessment (PANDA) test set. It exhibits superior generalization capacities when evaluated in more diverse clinical settings, achieving a quadratic kappa of 0.877, outperforming existing solutions. These results demonstrate our approach’s robustness and practical applicability, paving the way for its broader adoption in computational pathology and possibly other medical imaging tasks. Our code is publicly available at <span><span>https://github.com/computationalpathologygroup/hvit</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"105 ","pages":"Article 103663"},"PeriodicalIF":10.7,"publicationDate":"2025-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144587577","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}

{"title":"DIOR-ViT: Differential ordinal learning Vision Transformer for cancer classification in pathology images","authors":"Ju Cheon Lee ,&nbsp;Keunho Byeon ,&nbsp;Boram Song ,&nbsp;Kyungeun Kim ,&nbsp;Jin Tae Kwak","doi":"10.1016/j.media.2025.103708","DOIUrl":"10.1016/j.media.2025.103708","url":null,"abstract":"<div><div>In computational pathology, cancer grading has been mainly studied as a categorical classification problem, which does not utilize the ordering nature of cancer grades such as the higher the grade is, the worse the cancer is. To incorporate the ordering relationship among cancer grades, we introduce a differential ordinal learning problem in which we define and learn the degree of difference in the categorical class labels between pairs of samples by using their differences in the feature space. To this end, we propose a transformer-based neural network that simultaneously conducts both categorical classification and differential ordinal classification for cancer grading. We also propose a tailored loss function for differential ordinal learning. Evaluating the proposed method on three different types of cancer datasets, we demonstrate that the adoption of differential ordinal learning can improve the accuracy and reliability of cancer grading, outperforming conventional cancer grading approaches. The proposed approach should be applicable to other diseases and problems as they involve ordinal relationship among class labels.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"105 ","pages":"Article 103708"},"PeriodicalIF":10.7,"publicationDate":"2025-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144587715","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}