Medical image analysis最新文献

{"title":"Cellflow: Advancing pathological image augmentation from spatial views to temporal trajectories","authors":"Zeyu Liu ,&nbsp;Tianyi Zhang ,&nbsp;Yufang He ,&nbsp;Bo Wen ,&nbsp;Haoran Guo ,&nbsp;Peng Zhang ,&nbsp;Chenbin Ma ,&nbsp;Shangqing Lyu ,&nbsp;Yunlu Feng ,&nbsp;Yu Zhao ,&nbsp;Yueming Jin ,&nbsp;Dachun Zhao ,&nbsp;Guanglei Zhang","doi":"10.1016/j.media.2026.103995","DOIUrl":"10.1016/j.media.2026.103995","url":null,"abstract":"<div><div>Deep learning has advanced pathological image analysis but remains constrained by limited annotated data, especially for fine-grained diagnostic tasks such as tumor subtyping, grading, and cellularity assessment. While data augmentation alleviates this issue, existing methods are restricted to spatial manipulations that lack morphological plausibility and overlook the temporal attributes of pathological state transition. To address this gap, we propose <strong>Cellflow</strong>, the first temporal-aware generative framework for pathological image augmentation. Cellflow models pathological transition as smooth trajectories on a biological image manifold, generating intermediate states via a stair-based diffusion bridge with classifier-guided probability-flow ordinary differential equations. This design produces morphologically plausible sequences that capture both cellular details and tissue-level architecture. Evaluated on 7 diverse datasets across organs, staining modalities, and diagnostic tasks, Cellflow consistently outperforms 6 spatial augmentation methods and 4 state-of-the-art generative models, yielding improved classification performance, higher image fidelity, and preservation of temporal coherence. Quantitative cellularity analysis provides additional validation of the biological authenticity of transition sequences. By introducing temporal modeling into pathological data augmentation, Cellflow establishes a paradigm shift from spatial manipulations to biologically grounded temporal trajectories that advances robust model training, rare disease exploration, and educational simulation in computational pathology. The Code is available at <span><span>https://github.com/Rowerliu/Cellflow</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"110 ","pages":"Article 103995"},"PeriodicalIF":11.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146160398","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":"ESM-AnatTractNet: Advanced deep learning model of true positive eloquent white matter tractography to improve preoperative evaluation of pediatric epilepsy surgery","authors":"Min-Hee Lee ,&nbsp;Bohan Xiao ,&nbsp;Soumyanil Banerjee ,&nbsp;Hiroshi Uda ,&nbsp;Yoon Ho Hwang ,&nbsp;Csaba Juhász ,&nbsp;Eishi Asano ,&nbsp;Ming Dong ,&nbsp;Jeong-Won Jeong","doi":"10.1016/j.media.2026.103969","DOIUrl":"10.1016/j.media.2026.103969","url":null,"abstract":"<div><div>Accurate preoperative identification of true positive white matter pathways involved in critical eloquent functions such as motor, language, and vision plays a vital role in minimizing the risk of postoperative functional deficits and improving postoperative functional outcomes in pediatric epilepsy surgery. This study proposes a novel deep learning model: “ESM-AnatTractNet” that can accurately classify true positive eloquent white matter pathways across preoperative diffusion weighted imaging tractography data of 85 drug-resistant epilepsy patients (age: 10.70 ± 4.41 years). To enhance geometric and anatomical consistency of true positive tract classification, the ESM-AnatTractNet integrated two features in a point-cloud-based framework, 1) electro-physiologically confirmed spatial coordinates using electrical stimulation mapping (ESM) and 2) anatomically-contexted labels of the end-to-end neural connection using a standard brain atlas. Its overall performance was validated by accurately classifying 14 eloquent functional areas in whole brain, objectively optimizing resection margins to preserve eloquent functions using Kalman filter, and precisely predicting postoperative language outcomes using canonical correlation. Our ESM-AnatTractNet outperformed other baseline models, achieving an accuracy of 97% in correctly classifying eloquent areas within 10mm spatial resolution of clinical subdural grid electroencephalography. The Kalman filter analysis achieved 94% accuracy in predicting no deficits when the ESM-AnatTractNet-defined preservation zones were not resected. Postoperative decrease in language-related white matter connection efficacy defined by the ESM-AnatTractNet analysis was significantly associated with worse postoperative language outcome (R=0.73, p &lt; 0.001). Our findings demonstrate that the ESM-AnatTractNet improves non-invasive localization of true positive eloquent white matter pathways, supporting its potential to enhance current preoperative evaluation of pediatric epilepsy surgery.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"110 ","pages":"Article 103969"},"PeriodicalIF":11.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072191","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":"Revisiting lesion tracking in 3D total body photography","authors":"Wei-Lun Huang ,&nbsp;Minghao Xue ,&nbsp;Zhiyou Liu ,&nbsp;Davood Tashayyod ,&nbsp;Jun Kang ,&nbsp;Amir Gandjbakhche ,&nbsp;Misha Kazhdan ,&nbsp;Mehran Armand","doi":"10.1016/j.media.2026.103963","DOIUrl":"10.1016/j.media.2026.103963","url":null,"abstract":"<div><div>Melanoma is the most deadly form of skin cancer. Tracking the evolution of nevi and detecting new lesions across the body is essential for the early detection of melanoma. Despite prior work on longitudinal tracking of skin lesions in 3D total body photography, there are still several challenges, including 1) low accuracy for finding correct lesion pairs across scans, 2) sensitivity to noisy lesion detection, and 3) lack of large-scale datasets with numerous annotated lesion pairs. We propose a framework that takes in a pair of 3D textured meshes, matches lesions in the context of total body photography, and identifies unmatchable lesions. We start by computing correspondence maps bringing the source and target meshes to a template mesh. Using these maps to define source/target signals over the template domain, we construct a flow field aligning the mapped signals. The initial correspondence maps are then refined by advecting forward/backward along the flow field. Finally, lesion assignment is performed using the refined correspondence maps. We propose the first large-scale dataset for skin lesion tracking with 25K lesion pairs across 198 subjects. The proposed method achieves a success rate of 90.1% (at 10 mm criterion) for all pairs of annotated lesions and a matching accuracy of 98.1% for subjects with more than 200 lesions.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"110 ","pages":"Article 103963"},"PeriodicalIF":11.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072193","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":"VHU-Net: Variational hadamard U-Net for body MRI bias field correction","authors":"Xin Zhu ,&nbsp;Ahmet Enis Cetin ,&nbsp;Gorkem Durak ,&nbsp;Batuhan Gundogdu ,&nbsp;Ziliang Hong ,&nbsp;Hongyi Pan ,&nbsp;Ertugrul Aktas ,&nbsp;Elif Keles ,&nbsp;Hatice Savas ,&nbsp;Aytekin Oto ,&nbsp;Hiten Patel ,&nbsp;Adam B. Murphy ,&nbsp;Ashley Ross ,&nbsp;Frank Miller ,&nbsp;Baris Turkbey ,&nbsp;Ulas Bagci","doi":"10.1016/j.media.2026.103955","DOIUrl":"10.1016/j.media.2026.103955","url":null,"abstract":"<div><div>Bias field artifacts in magnetic resonance imaging (MRI) scans introduce spatially smooth intensity inhomogeneities that degrade image quality and hinder downstream analysis. To address this challenge, we propose a novel variational Hadamard U-Net (VHU-Net) for effective body MRI bias field correction. The encoder comprises multiple convolutional Hadamard transform blocks (ConvHTBlocks), each integrating convolutional layers with a Hadamard transform (HT) layer. Specifically, the HT layer performs channel-wise frequency decomposition to isolate low-frequency components, while a subsequent scaling layer and semi-soft thresholding mechanism suppress redundant high-frequency noise. To compensate for the HT layer’s inability to model inter-channel dependencies, the decoder incorporates an inverse HT-reconstructed transformer block, enabling global, frequency-aware attention for the recovery of spatially consistent bias fields. The stacked decoder ConvHTBlocks further enhance the capacity to reconstruct the underlying ground-truth bias field. Building on the principles of variational inference, we formulate a new evidence lower bound (ELBO) as the training objective, promoting sparsity in the latent space while ensuring accurate bias field estimation. Comprehensive experiments on body MRI datasets demonstrate the superiority of VHU-Net over existing state-of-the-art methods in terms of intensity uniformity. Moreover, the corrected images yield substantial downstream improvements in segmentation accuracy. Our framework offers computational efficiency, interpretability, and robust performance across multi-center datasets, making it suitable for clinical deployment. The codes are available at <span><span>https://github.com/Holmes696/Probabilistic-Hadamard-U-Net</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"110 ","pages":"Article 103955"},"PeriodicalIF":11.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146014237","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":"Slot-BERT: Self-supervised object discovery in surgical video","authors":"Guiqiu Liao ,&nbsp;Matjaž Jogan ,&nbsp;Marcel Hussing ,&nbsp;Kenta Nakahashi ,&nbsp;Kazuhiro Yasufuku ,&nbsp;Amin Madani ,&nbsp;Eric Eaton ,&nbsp;Daniel A. Hashimoto","doi":"10.1016/j.media.2026.103972","DOIUrl":"10.1016/j.media.2026.103972","url":null,"abstract":"<div><div>Object-centric slot attention is a powerful framework for unsupervised learning of structured and explainable representations that can support reasoning about objects and actions, including in surgical video. However, current object-centric models either fail to reliably capture object dependencies in seconds-long video episodes that encompass surgical actions and tasks or are computationally too expensive for practical implementation. We introduce Slot-BERT, a slot attention model with a temporal slot transformer module to overcome these limitations. Our core innovations are: 1) A bidirectional transformer module that processes object-centric slot representations, enabling longer-range temporal coherence; 2) A slot-contrastive loss that further improves the representation by enforcing slot dissimilarity; 3) We evaluate Slot-BERT on real-world surgical video datasets from abdominal, cholecystectomy, and thoracic procedures, and on real and synthetic videos with everyday objects. Our method surpasses state-of-the-art object-centric approaches under unsupervised training achieving superior performance across these domains. We also demonstrate efficient zero-shot domain adaptation to data from diverse surgical specialties and databases.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"110 ","pages":"Article 103972"},"PeriodicalIF":11.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146109926","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":"Robust fine-grained echocardiographic view classification with supervised contrastive learning","authors":"Preshen Naidoo ,&nbsp;Patricia Fernandes ,&nbsp;Nasim Dadashi Serej ,&nbsp;Catherine C Stowell ,&nbsp;Charlotte H Manisty ,&nbsp;Darrel P Francis ,&nbsp;Massoud Zolgharni","doi":"10.1016/j.media.2026.104006","DOIUrl":"10.1016/j.media.2026.104006","url":null,"abstract":"<div><div>Accurate classification of echocardiographic views is fundamental for automated cardiac analysis. However, clinical practice relies on a large, heterogeneous set of fine-grained acquisitions that introduce substantial inter-observer variability. Existing studies have primarily focused on limited view sets, often collapsing specialised views into broad categories, which limits their clinical relevance. We address this limitation by introducing TTE47, the first publicly available benchmark comprising 47 clinically meaningful views annotated independently by three experts. This dataset enables the rigorous quantification of inter-observer agreement and establishes a foundation for reproducible, clinically relevant evaluation. To tackle the dual challenges of subtle inter-class distinctions and structured label variability, we propose a novel supervised contrastive learning framework incorporating a tailored loss function. Our method outperforms cross-entropy and standard supervised contrastive baselines, achieving leading performance among evaluated methods on TTE47 and surpassing prior work on TMED-2 without dataset-specific pretraining, using a model pretrained on TTE47. Beyond accuracy, we introduce clustering-based metrics, Detection Rate and Label Recovery Precision, that measure semantic coherence and the model’s ability to resist annotation variability. Results show that the learned feature space aligns more strongly with underlying anatomical structure than with any single annotator’s style, enabling resilience to label shifts and maintaining robustness comparable to human-level disagreement. By integrating multi-expert evaluation, robust representation learning, and interpretable feature-space analysis, this work establishes a scalable and clinically relevant framework for fine-grained echo view classification. Our findings highlight the potential of contrastive pretraining to standardise interpretation, mitigate subjectivity, and enhance the reliability of AI-assisted echocardiography in diverse clinical settings.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"110 ","pages":"Article 104006"},"PeriodicalIF":11.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147334542","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":"Asymmetric fiber orientation distribution estimation via unsupervised deep learning","authors":"Di Zhang ,&nbsp;Ziyu Li ,&nbsp;Xiaofeng Deng ,&nbsp;Zekun Han ,&nbsp;Alan Wang ,&nbsp;Yong Liu ,&nbsp;Fangrong Zong","doi":"10.1016/j.media.2026.103968","DOIUrl":"10.1016/j.media.2026.103968","url":null,"abstract":"<div><div>Diffusion magnetic resonance imaging (dMRI) tractography is a key technique for reconstructing brain structural connectivity. A widely recognized limitation in tractography is the enforced symmetry of fiber orientation distribution functions (fODFs) in opposite directions, which may impair performance in regions with asymmetric microstructural organization. Previous studies have proposed leveraging anatomical priors or labeled training data to address this limitation; however, such data requirements constrain generalizability. In this study, we propose Recursive-a-fODF, a recursive estimator that uses an unsupervised deep learning framework to directly estimate asymmetric fODFs (a-fODFs) from dMRI data. The model incorporates a recursive calibration process that directly and dynamically estimates the white matter response function from the data itself, eliminating the need for external anatomical priors. We validate the framework using ex vivo marmoset brain data and in vivo human datasets, demonstrating superior performance in resolving complex fiber configurations. When applied to clinical cohorts with neurodegenerative and psychiatric conditions, Recursive-a-fODF reveals disease-specific alterations in fiber orientation asymmetry. These findings demonstrate that a-fODFs, estimated in a purely data-driven manner, can capture microstructural signatures relevant to disease pathology. Collectively, this work establishes a-fODF-based modeling as a powerful, anatomically unbiased approach that provides a complementary dimension to conventional diffusion metrics. These technical advances form a foundation for more accurate tractography and offer a new avenue for developing sensitive neuroimaging biomarkers.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"110 ","pages":"Article 103968"},"PeriodicalIF":11.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071491","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":"MADAT: Missing-aware dynamic adaptive transformer model for medical prognosis prediction with incomplete multimodal data","authors":"Jianbin He ,&nbsp;Guoheng Huang ,&nbsp;Xiaochen Yuan ,&nbsp;Chi-Man Pun ,&nbsp;Guo Zhong ,&nbsp;Qi Yang ,&nbsp;Ling Guo ,&nbsp;Siyu Zhu ,&nbsp;Baiying Lei ,&nbsp;Haojiang Li","doi":"10.1016/j.media.2026.103958","DOIUrl":"10.1016/j.media.2026.103958","url":null,"abstract":"<div><div>Multimodal medical prognosis prediction has shown great potential in improving diagnostic accuracy by integrating various data types. However, incomplete multimodality, where certain modalities are missing, poses significant challenges to model performance. Current methods, including dynamic adaptation and modality completion, have limitations in handling incomplete multimodality comprehensively. Dynamic adaptation methods fail to fully utilize modality interactions as they only process available modalities. Modality completion methods address inter-modal relationships but risk generating unreliable data, especially when key modalities are missing, since existing modalities cannot replicate unique features of absent ones. This compromises fusion quality and degrades model performance. To address these challenges, we propose the Missing-aware Dynamic Adaptive Transformer (MADAT) model, which integrates two phases: the Decoupling Generalization Completion Phase (DGCP), the Adaptive Cross-Fusion Phase (ACFP). The DGCP reconstructs missing modalities by generating inter-modal and intra-modal shared information using Progressive Transformation Recursive Gated Convolutions (PTRGC) and Wavelet Alignment Domain Generalization (WADG). The ACFP, which incorporates Cross-Agent Attention (CAA) and Generation Quality Feedback Regulation (GQFR), adaptively fuses the original and generated modality features. CAA ensures thorough integration and alignment of the features, while GQFR dynamically adjusts the model’s reliance on the generated features based on their quality, preventing over-dependence on low-quality data. Experiments on three private nasopharyngeal carcinoma datasets demonstrate that MADAT outperforms existing methods, achieving superior robustness in medical multimodal prediction under conditions of incomplete multimodality.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"110 ","pages":"Article 103958"},"PeriodicalIF":11.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146048256","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 foundation model for 3D magnetic resonance imaging segmentation, classification, and registration","authors":"Shansong Wang,&nbsp;Mojtaba Safari,&nbsp;Qiang Li,&nbsp;Chih-Wei Chang,&nbsp;Richard LJ Qiu,&nbsp;Justin Roper,&nbsp;David S. Yu,&nbsp;Xiaofeng Yang","doi":"10.1016/j.media.2026.103992","DOIUrl":"10.1016/j.media.2026.103992","url":null,"abstract":"<div><div>Vision foundation models (VFMs) are pre-trained on extensive image datasets to learn general representations. These models can subsequently be fine-tuned for specific downstream tasks, markedly boosting performance across a broad range of applications. However, existing vision foundation models that claim to be applicable to various downstream tasks are mostly pre-trained on imaging modalities with different characteristics than magnetic resonance imaging (MRI), those differences in imaging principles, signal characteristics, and data distribution may hinder their practical performance and versatility in MRI-specific applications. Here, we propose <strong>Triad</strong>, a vision foundation model for 3D MRI segmentation, classification, and registration. Triad learns robust representations from 129K 3D MRI volumes based on SimMIM framework and uses textual descriptions related to modality, device parameters, and imaging parameters to constrain the semantic distribution of the visual modality. The above pre-training dataset is called Triad-129K, which is currently the largest 3D MRI pre-training dataset. We evaluate Triad across three tasks, namely, organ/tumor segmentation, organ/cancer classification, and medical image registration, in two data modalities (within-domain and out-of-domain) settings using 25 downstream datasets. By initializing models with Triad’s pre-trained weights, nnUNet-Triad-SimMIM improves segmentation performance by 2.13% compared to nnUNet-Scratch across 17 datasets. Swin-B-Triad-SimMIM achieves a 4.38% improvement over Swin-B-Scratch in classification tasks across five datasets. SwinUNETR-Triad-SimMIM improves by 3.84% compared to SwinUNETR-Scratch in registration tasks across two datasets. Our study demonstrates that pre-training can improve performance when the data modalities and organs of upstream and downstream tasks are consistent. This work highlights the value of large-scale pre-training techniques for downstream tasks in 3D MRI.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"110 ","pages":"Article 103992"},"PeriodicalIF":11.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146209649","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":"Unsupervised domain adaptation for medical image segmentation using adaptogen-perturbation","authors":"Hong Joo Lee ,&nbsp;Yuan Bi ,&nbsp;Sangmin Lee ,&nbsp;Gyeong-Moon Park ,&nbsp;Jung Uk Kim ,&nbsp;Seong Tae Kim ,&nbsp;Zhongliang Jiang ,&nbsp;Nassir Navab","doi":"10.1016/j.media.2026.104002","DOIUrl":"10.1016/j.media.2026.104002","url":null,"abstract":"<div><div>Domains shift originated from differences in devices or patients in the medical field, poses a significant challenge when applying pre-trained models to clinical applications. To tackle this challenge, domain adaptation methods have been explored. However, most existing methods are designed for a single target domain adaptation or require sharing all target domain data for adaptation, which is infeasible in the medical field due to privacy issues. In this paper, we propose a novel unsupervised multi-target domain adaptation method without requiring data sharing. To this end, we introduce an additional signal, termed Adaptogen-Perturbation (AP) optimized to bridge the gap between the source and target domains. The optimized AP is injected into the latent feature and facilitates the adaptation of the pre-trained model to the target domain. Moreover, we propose a Spectral/Geometric Consistency learning framework to optimize the AP in an unsupervised manner. This promotes consistent predictions across two types of transformations: geometric and frequency-space spectral transformations, enhancing robustness to both variations. Extensive experiments with multiple medical segmentation datasets demonstrate the effectiveness of APs.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"110 ","pages":"Article 104002"},"PeriodicalIF":11.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146778369","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}