Medical image analysis最新文献

{"title":"Recurrent inference machine for medical image registration","authors":"Yi Zhang ,&nbsp;Yidong Zhao ,&nbsp;Hui Xue ,&nbsp;Peter Kellman ,&nbsp;Stefan Klein ,&nbsp;Qian Tao","doi":"10.1016/j.media.2025.103748","DOIUrl":"10.1016/j.media.2025.103748","url":null,"abstract":"<div><div>Image registration is essential for medical image applications where alignment of voxels across multiple images is needed for qualitative or quantitative analysis. With recent advances in deep neural networks and parallel computing, deep learning-based medical image registration methods become competitive with their flexible modeling and fast inference capabilities. However, compared to traditional optimization-based registration methods, the speed advantage may come at the cost of registration performance at inference time. Besides, deep neural networks ideally demand large training datasets while optimization-based methods are training-free. To improve registration accuracy and data efficiency, we propose a novel image registration method, termed Recurrent Inference Image Registration (RIIR) network. RIIR is formulated as a meta-learning solver for the registration problem in an iterative manner. RIIR addresses the accuracy and data efficiency issues, by learning the update rule of optimization, with implicit regularization combined with explicit gradient input.</div><div>We extensively evaluated RIIR on brain MRI, lung CT, and quantitative cardiac MRI datasets, in terms of both registration accuracy and training data efficiency. Our experiments showed that RIIR outperformed a range of deep learning-based methods, even with only 5% of the training data, demonstrating high data efficiency. Key findings from our ablation studies highlighted the important added value of the hidden states introduced in the recurrent inference framework for meta-learning. Our proposed RIIR offers a highly data-efficient framework for deep learning-based medical image registration.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103748"},"PeriodicalIF":11.8,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144804418","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":"Contrastive learning and prior knowledge-induced feature extraction network for prediction of high-risk recurrence areas in Gliomas","authors":"Boya Wu ,&nbsp;Jianyun Cao ,&nbsp;Wei Xiong ,&nbsp;Yanchun Lv ,&nbsp;Guohua Zhao ,&nbsp;Xiaoyue Ma ,&nbsp;Ying Zhang ,&nbsp;Jiawei Zhang ,&nbsp;Junguo Bu ,&nbsp;Tao Xie ,&nbsp;Qianjin Feng ,&nbsp;Meiyan Huang","doi":"10.1016/j.media.2025.103740","DOIUrl":"10.1016/j.media.2025.103740","url":null,"abstract":"<div><div>Gliomas can easily recur even after standard treatments, and their recurrence may be related to insufficient radiation doses received by high-risk recurrence areas (HRA). Therefore, HRA prediction can help clinical experts in formulating effective radiotherapy plans. However, research on HRA prediction using early postoperative conventional MRI images with total resection is lacking. This gap is due to multifold challenges, including visually minimal differences between HRA and non-HRA and small dataset size caused by missing follow-up data. A contrastive learning and prior knowledge-induced feature extraction network (CLPKnet) to explore HRA-related features and achieve HRA prediction was proposed in this paper. First, a contrastive and multisequence learning-based encoder was proposed to effectively extract diverse features across multiple MRI sequences around the operative cavity. Specifically, a contrastive learning method was employed to pretrain the encoder, which enabled it to capture subtle differences between HRA and non-HRA regions while mitigating the challenges posed by the limited dataset size. Second, clinical prior knowledge was incorporated into the CLPKnet to guide the model in learning the patterns of glioma growth and improve its discriminative capability for identifying HRA regions. Third, a dual-focus fusion module was utilized to explore important sequential features and spatial regions and effectively fused multisequence features to provide complementary information associated with glioma recurrence. Fourth, to balance clinical needs and task difficulty, we used a patch-based prediction method to predict the recurrent probability. The CLPKnet was validated on a multicenter dataset from four hospitals, and a remarkable performance was achieved. Moreover, the interpretability and robustness of our method were evaluated to illustrate its effectiveness and credibility. Therefore, the CLPKnet displays a great application potential for HRA prediction. The codes will be available at <span><span>https://github.com/Meiyan88/CLPKnet</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103740"},"PeriodicalIF":11.8,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144780232","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":"Interpretable multi-scale deep learning to detect malignancy in cell blocks and cytological smears of pleural effusion and identify aggressive endometrial cancer","authors":"Ching-Wei Wang ,&nbsp;Hikam Muzakky ,&nbsp;Yu-Pang Chung ,&nbsp;Po-Jen Lai ,&nbsp;Tai-Kuang Chao","doi":"10.1016/j.media.2025.103742","DOIUrl":"10.1016/j.media.2025.103742","url":null,"abstract":"<div><div>The pleura is a serous membrane that surrounds the surface of the lungs. The visceral surface secretes fluid into the serous cavity, while the parietal surface ensures that the fluid is properly absorbed. However, when this balance is disrupted, it leads to the formation of pleural Effusion. The most common malignant pleural effusion (MPE) caused by lung cancer or breast cancer, and benign pleural effusions (BPE) caused by Mycobacterium tuberculosis infection, heart failure, or infections related to pneumonia. Today, with the rapid advancement of treatment protocols, accurately diagnosing MPE has become increasingly important. Although cytology smears and cell blocks examinations of pleural effusion are the clinical gold standards for diagnosing MPE, the diagnostic accuracy of these tools can be affected by certain limitations, such as low sensitivity, diagnostic variability across different regions and significant inter-observer variability, leading to a certain proportion of misdiagnoses. This study presents a deep learning (DL) framework, namely Interpretable Multi-scale Attention DL with Self-Supervised Learning Feature Encoder (IMA-SSL), to identifyMPE or BPE using 194 Cytological smears whole-slide images (WSIs) and 188 cell blocks WSIs. The use of DL on WSIs of pleural effusion allows for preliminary results to be obtained in a short time, giving patients the opportunity for earlier diagnosis and treatment. The experimental results show that the proposed IMA-SSL consistently obtained superior performance and outperformed five state-of-the-art (SOTA) methods in malignancy prediction on both cell block and cytological smear datasets and also in identification of aggressive endometrial cancer (EC) using a public TCGA dataset. Fisher’s exact test confirmed a highly significant correlation between the outputs of the proposed model and the slide status in the EC and pleural effusion datasets (<span><math><mi>p &lt; 0.001</mi></math></span>), substantiating the model’s predictive reliability. The proposed method has the potential for practical clinical application in the foreseeable future. It can directly detect the presence of malignant tumor cells from cost-effective cell blocks and pleural effusion cytology smears and facilitate personalized cancer treatment decisions.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103742"},"PeriodicalIF":11.8,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144804415","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":"MedCLIP-SAMv2: Towards universal text-driven medical image segmentation","authors":"Taha Koleilat ,&nbsp;Hojat Asgariandehkordi ,&nbsp;Hassan Rivaz ,&nbsp;Yiming Xiao","doi":"10.1016/j.media.2025.103749","DOIUrl":"10.1016/j.media.2025.103749","url":null,"abstract":"<div><div>Segmentation of anatomical structures and pathologies in medical images is essential for modern disease diagnosis, clinical research, and treatment planning. While significant advancements have been made in deep learning-based segmentation techniques, many of these methods still suffer from limitations in data efficiency, generalizability, and interactivity. As a result, developing robust segmentation methods that require fewer labeled datasets remains a critical challenge in medical image analysis. Recently, the introduction of foundation models like CLIP and Segment-Anything-Model (SAM), with robust cross-domain representations, has paved the way for interactive and universal image segmentation. However, further exploration of these models for data-efficient segmentation in medical imaging is an active field of research. In this paper, we introduce MedCLIP-SAMv2, a novel framework that integrates the CLIP and SAM models to perform segmentation on clinical scans using text prompts, in both zero-shot and weakly supervised settings. Our approach includes fine-tuning the BiomedCLIP model with a new Decoupled Hard Negative Noise Contrastive Estimation (DHN-NCE) loss, and leveraging the Multi-modal Information Bottleneck (M2IB) to create visual prompts for generating segmentation masks with SAM in the zero-shot setting. We also investigate using zero-shot segmentation labels in a weakly supervised paradigm to enhance segmentation quality further. Extensive validation across four diverse segmentation tasks and medical imaging modalities (breast tumor ultrasound, brain tumor MRI, lung X-ray, and lung CT) demonstrates the high accuracy of our proposed framework. Our code is available at <span><span>https://github.com/HealthX-Lab/MedCLIP-SAMv2</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103749"},"PeriodicalIF":11.8,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144804417","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":"Brain Latent Progression: Individual-based spatiotemporal disease progression on 3D Brain MRIs via latent diffusion","authors":"Lemuel Puglisi ,&nbsp;Daniel C. Alexander ,&nbsp;Alzheimer’s Disease Neuroimaging Initiative ,&nbsp;Australian Imaging Biomarkers and Lifestyle flagship study of aging ,&nbsp;Daniele Ravì","doi":"10.1016/j.media.2025.103734","DOIUrl":"10.1016/j.media.2025.103734","url":null,"abstract":"<div><div>The growing availability of longitudinal Magnetic Resonance Imaging (MRI) datasets has facilitated Artificial Intelligence (AI)-driven modeling of disease progression, making it possible to predict future medical scans for individual patients. However, despite significant advancements in AI, current methods continue to face challenges including achieving patient-specific individualization, ensuring spatiotemporal consistency, efficiently utilizing longitudinal data, and managing the substantial memory demands of 3D scans. To address these challenges, we propose Brain Latent Progression (BrLP), a novel spatiotemporal model designed to predict individual-level disease progression in 3D brain MRIs. The key contributions in BrLP are fourfold: (i) it operates in a small latent space, mitigating the computational challenges posed by high-dimensional imaging data; (ii) it explicitly integrates subject metadata to enhance the individualization of predictions; (iii) it incorporates prior knowledge of disease dynamics through an auxiliary model, facilitating the integration of longitudinal data; and (iv) it introduces the Latent Average Stabilization (LAS) algorithm, which (a) enforces spatiotemporal consistency in the predicted progression at inference time and (b) allows us to derive a measure of the uncertainty for the prediction at the global and voxel level. We train and evaluate BrLP on 11,730 T1-weighted (T1w) brain MRIs from 2,805 subjects and validate its generalizability on an external test set comprising 2,257 MRIs from 962 subjects. Our experiments compare BrLP-generated MRI scans with real follow-up MRIs, demonstrating state-of-the-art accuracy compared to existing methods. The code is publicly available at: <span><span>https://github.com/LemuelPuglisi/BrLP</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103734"},"PeriodicalIF":11.8,"publicationDate":"2025-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144780233","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":"A generalizable diffusion framework for 3D low-dose and few-view cardiac SPECT imaging","authors":"Huidong Xie ,&nbsp;Weijie Gan ,&nbsp;Wei Ji ,&nbsp;Xiongchao Chen ,&nbsp;Alaa Alashi ,&nbsp;Stephanie L. Thorn ,&nbsp;Bo Zhou ,&nbsp;Qiong Liu ,&nbsp;Menghua Xia ,&nbsp;Xueqi Guo ,&nbsp;Yi-Hwa Liu ,&nbsp;Hongyu An ,&nbsp;Ulugbek S. Kamilov ,&nbsp;Ge Wang ,&nbsp;Albert J. Sinusas ,&nbsp;Chi Liu","doi":"10.1016/j.media.2025.103729","DOIUrl":"10.1016/j.media.2025.103729","url":null,"abstract":"<div><div>Myocardial perfusion imaging using SPECT is widely utilized to diagnose coronary artery diseases, but image quality can be negatively affected in low-dose and few-view acquisition settings. Although various deep learning methods have been introduced to improve image quality from low-dose or few-view SPECT data, previous approaches often fail to generalize across different acquisition settings, limiting realistic applicability. This work introduced DiffSPECT-3D, a diffusion framework for 3D cardiac SPECT imaging that effectively adapts to different acquisition settings without requiring further network re-training or fine-tuning. Using both image and projection data, a consistency strategy is proposed to ensure that diffusion sampling at each step aligns with the low-dose/few-view projection measurements, the image data, and the scanner geometry, thus enabling generalization to different low-dose/few-view settings. Incorporating anatomical spatial information from CT and total variation constraint, we proposed a 2.5D conditional strategy to allow DiffSPECT-3D to observe 3D contextual information from the entire image volume, addressing the 3D memory/computational issues in diffusion model. We extensively evaluated the proposed method on 1,325 clinical <span><math><msup><mrow></mrow><mrow><mtext>99m</mtext></mrow></msup></math></span>Tc tetrofosmin stress/rest studies from 795 patients. Each study was reconstructed into 5 different low-count levels and 5 different projection few-view levels for model evaluations, ranging from 1% to 50% and from 1 view to 9 view, respectively. Validated against cardiac catheterization results and diagnostic review from nuclear cardiologists, the presented results show the potential to achieve low-dose and few-view SPECT imaging without compromising clinical performance. Additionally, DiffSPECT-3D could be directly applied to full-dose SPECT images to further improve image quality, especially in a low-dose stress-first cardiac SPECT imaging protocol.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103729"},"PeriodicalIF":11.8,"publicationDate":"2025-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144768766","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":"Pixel-wise recognition for holistic surgical scene understanding","authors":"Nicolás Ayobi ,&nbsp;Santiago Rodríguez ,&nbsp;Alejandra Pérez ,&nbsp;Isabela Hernández ,&nbsp;Nicolás Aparicio ,&nbsp;Eugénie Dessevres ,&nbsp;Sebastián Peña ,&nbsp;Jessica Santander ,&nbsp;Juan Ignacio Caicedo ,&nbsp;Nicolás Fernández ,&nbsp;Pablo Arbeláez","doi":"10.1016/j.media.2025.103726","DOIUrl":"10.1016/j.media.2025.103726","url":null,"abstract":"<div><div>This paper presents the Holistic and Multi-Granular Surgical Scene Understanding of Prostatectomies (GraSP) dataset, a curated benchmark that models surgical scene understanding as a hierarchy of complementary tasks with varying levels of granularity. Our approach encompasses long-term tasks, such as surgical phase and step recognition, and short-term tasks, including surgical instrument segmentation and atomic visual actions detection. To exploit our proposed benchmark, we introduce the Transformers for Actions, Phases, Steps, and Instrument Segmentation (TAPIS) model, a general architecture that combines a global video feature extractor with localized region proposals from an instrument segmentation model to tackle the multi-granularity of our benchmark. We demonstrate TAPIS’s versatility and state-of-the-art performance across different tasks through extensive experimentation on GraSP and alternative benchmarks. This work represents a foundational step forward in Endoscopic Vision, offering a novel framework for future research towards holistic surgical scene understanding.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"107 ","pages":"Article 103726"},"PeriodicalIF":11.8,"publicationDate":"2025-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144799560","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":"Synomaly noise and multi-stage diffusion: A novel approach for unsupervised anomaly detection in medical images","authors":"Yuan Bi ,&nbsp;Lucie Huang ,&nbsp;Ricarda Clarenbach ,&nbsp;Reza Ghotbi ,&nbsp;Angelos Karlas ,&nbsp;Nassir Navab ,&nbsp;Zhongliang Jiang","doi":"10.1016/j.media.2025.103737","DOIUrl":"10.1016/j.media.2025.103737","url":null,"abstract":"<div><div>Anomaly detection in medical imaging plays a crucial role in identifying pathological regions across various imaging modalities, such as brain MRI, liver CT, and carotid ultrasound (US). However, training fully supervised segmentation models is often hindered by the scarcity of expert annotations and the complexity of diverse anatomical structures. To address these issues, we propose a novel unsupervised anomaly detection framework based on a diffusion model that incorporates a synthetic anomaly (Synomaly) noise function and a multi-stage diffusion process. Synomaly noise introduces synthetic anomalies into healthy images during training, allowing the model to effectively learn anomaly removal. The multi-stage diffusion process is introduced to progressively denoise images, preserving fine details while improving the quality of anomaly-free reconstructions. The generated high-fidelity counterfactual healthy images can further enhance the interpretability of the segmentation models, as well as provide a reliable baseline for evaluating the extent of anomalies and supporting clinical decision-making. Notably, the unsupervised anomaly detection model is trained purely on healthy images, eliminating the need for anomalous training samples and pixel-level annotations. We validate the proposed approach on brain MRI, liver CT datasets, and carotid US. The experimental results demonstrate that the proposed framework outperforms existing state-of-the-art unsupervised anomaly detection methods, achieving performance comparable to fully supervised segmentation models in the US dataset. Ablation studies further highlight the contributions of Synomaly noise and the multi-stage diffusion process in improving anomaly segmentation. These findings underscore the potential of our approach as a robust and annotation-efficient alternative for medical anomaly detection. <strong>Code:</strong> <span><span>https://github.com/yuan-12138/Synomaly</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"105 ","pages":"Article 103737"},"PeriodicalIF":11.8,"publicationDate":"2025-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144738204","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":"Accelerating cardiac radial-MRI: Fully polar based technique using compressed sensing and deep learning","authors":"Vahid Ghodrati ,&nbsp;Jinming Duan ,&nbsp;Fadil Ali ,&nbsp;Arash Bedayat ,&nbsp;Ashley Prosper ,&nbsp;Mark Bydder","doi":"10.1016/j.media.2025.103732","DOIUrl":"10.1016/j.media.2025.103732","url":null,"abstract":"<div><div>Fast radial-MRI approaches based on compressed sensing (CS) and deep learning (DL) often use non-uniform fast Fourier transform (NUFFT) as the forward imaging operator, which might introduce interpolation errors and reduce image quality. Using the polar Fourier transform (PFT), we developed fully polar CS and DL algorithms for fast 2D cardiac radial-MRI. Our methods directly reconstruct images in polar spatial space from polar k-space data, eliminating frequency interpolation and ensuring an easy-to-compute data consistency term for the DL framework via the variable splitting (VS) scheme. Furthermore, PFT reconstruction produces initial images with fewer artifacts in a reduced field of view, making it a better starting point for CS and DL algorithms, especially for dynamic imaging, where information from a small region of interest is critical, as opposed to NUFFT, which often results in global streaking artifacts. In the cardiac region, PFT-based CS technique outperformed NUFFT-based CS at acceleration rates of 5x (mean SSIM: 0.8831 vs. 0.8526), 10x (0.8195 vs. 0.7981), and 15x (0.7720 vs. 0.7503). Our PFT(VS)-DL technique outperformed the NUFFT(GD)-based DL method, which used unrolled gradient descent with the NUFFT as the forward imaging operator, with mean SSIM scores of 0.8914 versus 0.8617 at 10x and 0.8470 versus 0.8301 at 15x. Radiological assessments revealed that PFT(VS)-based DL scored <span><math><mrow><mn>2</mn><mo>.</mo><mn>9</mn><mo>±</mo><mn>0</mn><mo>.</mo><mn>30</mn></mrow></math></span> and <span><math><mrow><mn>2</mn><mo>.</mo><mn>73</mn><mo>±</mo><mn>0</mn><mo>.</mo><mn>45</mn></mrow></math></span> at 5x and 10x, whereas NUFFT(GD)-based DL scored <span><math><mrow><mn>2</mn><mo>.</mo><mn>7</mn><mo>±</mo><mn>0</mn><mo>.</mo><mn>47</mn></mrow></math></span> and <span><math><mrow><mn>2</mn><mo>.</mo><mn>40</mn><mo>±</mo><mn>0</mn><mo>.</mo><mn>50</mn></mrow></math></span>, respectively. Our methods suggest a promising alternative to NUFFT-based fast radial-MRI for dynamic imaging, prioritizing reconstruction quality in a small region of interest over whole image quality.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"105 ","pages":"Article 103732"},"PeriodicalIF":11.8,"publicationDate":"2025-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144738205","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":"Controllable illumination invariant GAN for diverse temporally-consistent surgical video synthesis","authors":"Long Chen ,&nbsp;Mobarak I. Hoque ,&nbsp;Zhe Min ,&nbsp;Matt Clarkson ,&nbsp;Thomas Dowrick","doi":"10.1016/j.media.2025.103731","DOIUrl":"10.1016/j.media.2025.103731","url":null,"abstract":"<div><div>Surgical video synthesis offers a cost-effective way to expand training data and enhance the performance of machine learning models in computer-assisted surgery. However, existing video translation methods often produce video sequences with large illumination changes across different views, disrupting the temporal consistency of the videos. Additionally, these methods typically synthesize videos with a monotonous style, whereas diverse synthetic data is desired to improve the generalization ability of downstream machine learning models. To address these challenges, we propose a novel Controllable Illumination Invariant Generative Adversarial Network (CIIGAN) for generating diverse, illumination-consistent video sequences. CIIGAN fuses multi-scale illumination-invariant features from a novel controllable illumination-invariant (CII) image space with multi-scale texture-invariant features from self-constructed 3D scenes. The CII image space, along with the 3D scenes, allows CIIGAN to produce diverse and temporally-consistent video or image translations. Extensive experiments demonstrate that CIIGAN achieves more realistic and illumination-consistent translations compared to previous state-of-the-art baselines. Furthermore, the segmentation networks trained on our diverse synthetic data outperform those trained on monotonous synthetic data. Our source code, well-trained models, and 3D simulation scenes are public available at <span><span>https://github.com/LongChenCV/CIIGAN</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"105 ","pages":"Article 103731"},"PeriodicalIF":10.7,"publicationDate":"2025-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144713306","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}