Pub Date : 2025-01-01DOI: 10.1016/j.bbe.2024.11.001
Haoneng Lin , Jing Zou , Sen Deng , Ka Po Wong , Angelica I. Aviles-Rivero , Yiting Fan , Alex Pui-Wai Lee , Xiaowei Hu , Jing Qin
The Segment Anything Model (SAM) exhibits exceptional generalization capabilities in diverse domains, owing to its interactive learning mechanism designed for precise image segmentation. However, applying SAM to out-of-distribution domains, especially in 3D medical image segmentation, poses challenges. Existing methods for adapting 2D segmentation models to 3D medical data treat 3D volumes as a mere stack of 2D slices. The essential inter-slice information, which is pivotal to faithful 3D medical image segmentation tasks, is unfortunately neglected. In this work, we present the 3D Medical SAM-Adapter (3DMedSAM), a pioneer cross-dimensional adaptation, leveraging SAM’s pre-trained knowledge while accommodating the unique characteristics of 3D medical data. Firstly, to bridge the dimensional gap from 2D to 3D, we design a novel module to replace SAM’s patch embedding, ensuring a seamless transition into 3D image processing and recognition. Besides, we incorporate a 3D Adapter while maintaining the majority of pre-training parameters frozen, enriching deep features with abundant 3D spatial information and achieving efficient fine-tuning. Given the diverse scales of anomalies present in medical images, we also devised a multi-scale 3D mask decoder to elevate the network’s proficiency in medical image segmentation. Through various experiments, we showcase the effectiveness of 3DMedSAM in achieving accurate and robust 3D segmentation on both single-target segmentation and multi-organ segmentation tasks, surpassing the limitations of current methods.
{"title":"Volumetric medical image segmentation via fully 3D adaptation of Segment Anything Model","authors":"Haoneng Lin , Jing Zou , Sen Deng , Ka Po Wong , Angelica I. Aviles-Rivero , Yiting Fan , Alex Pui-Wai Lee , Xiaowei Hu , Jing Qin","doi":"10.1016/j.bbe.2024.11.001","DOIUrl":"10.1016/j.bbe.2024.11.001","url":null,"abstract":"<div><div>The Segment Anything Model (SAM) exhibits exceptional generalization capabilities in diverse domains, owing to its interactive learning mechanism designed for precise image segmentation. However, applying SAM to out-of-distribution domains, especially in 3D medical image segmentation, poses challenges. Existing methods for adapting 2D segmentation models to 3D medical data treat 3D volumes as a mere stack of 2D slices. The essential inter-slice information, which is pivotal to faithful 3D medical image segmentation tasks, is unfortunately neglected. In this work, we present the 3D Medical SAM-Adapter (3DMedSAM), a pioneer cross-dimensional adaptation, leveraging SAM’s pre-trained knowledge while accommodating the unique characteristics of 3D medical data. Firstly, to bridge the dimensional gap from 2D to 3D, we design a novel module to replace SAM’s patch embedding, ensuring a seamless transition into 3D image processing and recognition. Besides, we incorporate a 3D Adapter while maintaining the majority of pre-training parameters frozen, enriching deep features with abundant 3D spatial information and achieving efficient fine-tuning. Given the diverse scales of anomalies present in medical images, we also devised a multi-scale 3D mask decoder to elevate the network’s proficiency in medical image segmentation. Through various experiments, we showcase the effectiveness of 3DMedSAM in achieving accurate and robust 3D segmentation on both single-target segmentation and multi-organ segmentation tasks, surpassing the limitations of current methods.</div></div>","PeriodicalId":55381,"journal":{"name":"Biocybernetics and Biomedical Engineering","volume":"45 1","pages":"Pages 1-10"},"PeriodicalIF":5.3,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143093056","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-01DOI: 10.1016/j.bbe.2025.01.004
Alireza Karimi , Marie Darche , Ansel Stanik , Reza Razaghi , Iman Mirafzal , Kamran Hassani , Mojtaba Hassani , Elizabeth White , Ivana Gantar , Stéphane Pagès , Laura Batti , Ted S. Acott , Michel Paques
Objective
Aging results in significant structural and functional changes in the anterior segment of the eye, influencing intraocular pressure (IOP) and overall ocular health. Although aging is a well-established risk factor for primary open-angle glaucoma, a leading cause of irreversible blindness, the specific mechanisms through which aging drives morphological changes in anterior segment tissues and affects aqueous humor dynamics remain incompletely understood.
Methods
In this study, we employed cutting-edge light sheet fluorescence microscopy (LSFM) to capture high-resolution, volumetric images of cleared human donor eyes’ anterior segment tissues. This advanced imaging enabled a comprehensive morphological analysis of key parameters, including central and peripheral corneal thickness (CCT and PCT), iris thickness, anterior chamber area (ACA), and ciliary body area (CBA). By integrating these morphological parameters with computational fluid dynamics (CFD) models, we analyzed aqueous humor dynamics across n = 6 female human donor eyes, spanning a wide age range of 5 to 94 years (all of Caucasian descent).
Results
The CCT and PCT demonstrated thinning with age, accompanied by a reduction in ACA. In contrast, the CBA remained relatively stable across all age groups. Computational fluid dynamics analysis showed a decline in aqueous humor velocity and wall shear stress, with younger eyes exhibiting higher velocities and shear stress, compared to older eyes.
Conclusion
These findings emphasize the value of integrating LSFM and CFD approaches to provide a detailed understanding of how aging impacts the anterior segment and its fluid dynamics. This study contributes to the understanding of age-related ocular changes, highlighting the importance of considering these changes in the diagnosis and management of age-related eye diseases.
{"title":"Impact of aging on anterior segment morphology and aqueous humor dynamics in human Eyes: Advanced imaging and computational techniques","authors":"Alireza Karimi , Marie Darche , Ansel Stanik , Reza Razaghi , Iman Mirafzal , Kamran Hassani , Mojtaba Hassani , Elizabeth White , Ivana Gantar , Stéphane Pagès , Laura Batti , Ted S. Acott , Michel Paques","doi":"10.1016/j.bbe.2025.01.004","DOIUrl":"10.1016/j.bbe.2025.01.004","url":null,"abstract":"<div><h3>Objective</h3><div>Aging results in significant structural and functional changes in the anterior segment of the eye, influencing intraocular pressure (IOP) and overall ocular health. Although aging is a well-established risk factor for primary open-angle glaucoma, a leading cause of irreversible blindness, the specific mechanisms through which aging drives morphological changes in anterior segment tissues and affects aqueous humor dynamics remain incompletely understood.</div></div><div><h3>Methods</h3><div>In this study, we employed cutting-edge light sheet fluorescence microscopy (LSFM) to capture high-resolution, volumetric images of cleared human donor eyes’ anterior segment tissues. This advanced imaging enabled a comprehensive morphological analysis of key parameters, including central and peripheral corneal thickness (CCT and PCT), iris thickness, anterior chamber area (ACA), and ciliary body area (CBA). By integrating these morphological parameters with computational fluid dynamics (CFD) models, we analyzed aqueous humor dynamics across <em>n</em> = 6 female human donor eyes, spanning a wide age range of 5 to 94 years (all of Caucasian descent).</div></div><div><h3>Results</h3><div>The CCT and PCT demonstrated thinning with age, accompanied by a reduction in ACA. In contrast, the CBA remained relatively stable across all age groups. Computational fluid dynamics analysis showed a decline in aqueous humor velocity and wall shear stress, with younger eyes exhibiting higher velocities and shear stress, compared to older eyes.</div></div><div><h3>Conclusion</h3><div>These findings emphasize the value of integrating LSFM and CFD approaches to provide a detailed understanding of how aging impacts the anterior segment and its fluid dynamics. This study contributes to the understanding of age-related ocular changes, highlighting the importance of considering these changes in the diagnosis and management of age-related eye diseases.</div></div>","PeriodicalId":55381,"journal":{"name":"Biocybernetics and Biomedical Engineering","volume":"45 1","pages":"Pages 62-73"},"PeriodicalIF":5.3,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143093057","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号