{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,8,2]],"date-time":"2025-08-02T16:51:27Z","timestamp":1754153487780,"version":"3.41.2"},"reference-count":43,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,7,23]],"date-time":"2025-07-23T00:00:00Z","timestamp":1753228800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100003848","name":"Industrial Technology Research Institute","doi-asserted-by":"publisher","award":["Q301AA3110"],"award-info":[{"award-number":["Q301AA3110"]}],"id":[{"id":"10.13039\/501100003848","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:sec><jats:title>Background<\/jats:title><jats:p>Abdominal ultrasonography is a primary diagnostic tool for evaluating medical conditions within the abdominal cavity. Accurate determination of the relative locations of intra-abdominal organs and lesions based on anatomical features in ultrasound images is essential in diagnostic sonography. Recognizing and extracting anatomical landmarks facilitates lesion evaluation and enhances diagnostic interpretation. Recent artificial intelligence (AI) segmentation methods employing deep neural networks (DNNs) and transformers encounter computational efficiency challenges to balance the preservation of feature dependencies information with model efficiency, limiting their clinical applicability.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>The anatomical structure recognition framework, MaskHybrid, was developed using a private dataset comprising 34,711 abdominal ultrasound images of 2,063 patients from CSMUH. The dataset included abdominal organs and vascular structures (hepatic vein, inferior vena cava, portal vein, gallbladder, kidney, pancreas, spleen) and liver lesions (hepatic cyst, tumor). MaskHybrid adopted a mamba-transformer hybrid architecture consisting of an evolved backbone network, encoder, and corresponding decoder to capture long-range spatial dependencies and contextual information effectively, demonstrating improved image segmentation capabilities in visual tasks while mitigating the computational burden associated with the transformer-based attention mechanism.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Experiments on the retrospective dataset achieved a mean average precision (mAP) score of 74.13% for anatomical landmarks segmentation in abdominal ultrasound images. Our proposed framework outperformed baselines across most organ and lesion types and effectively segmented challenging anatomical structures. Moreover, MaskHybrid exhibited a significantly shorter inference time (0.120\u202f\u00b1\u202f0.013\u202fs), achieving 2.5 times faster than large-sized AI models of similar size. Combining Mamba and transformer architectures, this hybrid design was well-suited for the timely analysis of complex anatomical structures segmentation in abdominal ultrasonography, where accuracy and efficiency are critical in clinical practice.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>The proposed mamba-transformer hybrid recognition framework simultaneously detects and segments multiple abdominal organs and lesions in ultrasound images, achieving superior segmentation accuracy, visualization effect, and inference efficiency, thereby facilitating improved medical image interpretation and near real-time diagnostic sonography that meets clinical needs.<\/jats:p><\/jats:sec>","DOI":"10.3389\/frai.2025.1618607","type":"journal-article","created":{"date-parts":[[2025,7,23]],"date-time":"2025-07-23T05:36:41Z","timestamp":1753249001000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["AI-assisted anatomical structure recognition and segmentation via mamba-transformer architecture in abdominal ultrasound images"],"prefix":"10.3389","volume":"8","author":[{"given":"Shih-Fang","family":"Chang","sequence":"first","affiliation":[]},{"given":"Po-Yi","family":"Wu","sequence":"additional","affiliation":[]},{"given":"Ming-Chang","family":"Tsai","sequence":"additional","affiliation":[]},{"given":"Vincent S.","family":"Tseng","sequence":"additional","affiliation":[]},{"given":"Chi-Chih","family":"Wang","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2025,7,23]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"165","DOI":"10.1016\/j.cmpb.2017.12.016","article-title":"Symtosis: a liver ultrasound tissue characterization and risk stratification in optimized deep learning paradigm","volume":"155","author":"Biswas","year":"2018","journal-title":"Comput. Methods Prog. Biomed."},{"year":"2024","author":"Boesch","key":"ref2"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"1775","DOI":"10.1007\/s00261-024-04640-x","article-title":"Artificial intelligence in abdominal and pelvic ultrasound imaging: current applications","volume":"50","author":"Cai","year":"2025","journal-title":"Abdom. Radiol."},{"year":"2020","author":"Carion","key":"ref4"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"105118","DOI":"10.1016\/j.cmpb.2019.105118","article-title":"Computer-aided diagnosis of gallbladder polyps based on high resolution ultrasonography","volume":"185","author":"Chen","year":"2020","journal-title":"Comput. Methods Prog. Biomed."},{"key":"ref6","doi-asserted-by":"publisher","first-page":"234","DOI":"10.1007\/s10278-016-9929-2","article-title":"Transfer learning with convolutional neural networks for classification of abdominal ultrasound images","volume":"30","author":"Cheng","year":"2017","journal-title":"J. Digit. Imaging"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"110","DOI":"10.1148\/ryai.210110","article-title":"Deep learning for the detection, localization, and characterization of focal liver lesions on abdominal US images","volume":"4","author":"Dadoun","year":"2022","journal-title":"Radiol. Artif. Intell."},{"year":"2020","author":"Dandan","key":"ref8"},{"year":"2024","author":"Dao","key":"ref9"},{"year":"2021","author":"Ghiasi","key":"ref10"},{"year":"2024","author":"Gu","key":"ref11"},{"year":"2024","author":"Hatamizadeh","key":"ref12"},{"year":"2021","author":"Hatture","key":"ref13"},{"year":"2017","author":"He","key":"ref14"},{"year":"2016","author":"He","key":"ref15"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"186","DOI":"10.1016\/j.media.2019.07.005","article-title":"Accurate and robust deep learning-based segmentation of the prostate clinical target volume in ultrasound images","volume":"57","author":"Karimi","year":"2019","journal-title":"Med. Image Anal."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"530","DOI":"10.1049\/iet-ipr.2016.1072","article-title":"Hybrid approach to classification of focal and diffused liver disorders using ultrasound images with wavelets and texture features","volume":"11","author":"Krishnan","year":"2017","journal-title":"IET Image Process."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"1264","DOI":"10.1007\/s00330-019-06407-1","article-title":"Deep learning with ultrasonography: automated classification of liver fibrosis using a deep convolutional neural network","volume":"30","author":"Lee","year":"2020","journal-title":"Eur. Radiol."},{"key":"ref19","doi-asserted-by":"publisher","first-page":"3194","DOI":"10.1002\/mp.13577","article-title":"Ultrasound prostate segmentation based on multidirectional deeply supervised V-net","volume":"46","author":"Lei","year":"2019","journal-title":"Med. Phys."},{"year":"2023","author":"Li","key":"ref20"},{"year":"2021","author":"Liu","key":"ref21"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"1877","DOI":"10.3390\/life12111877","article-title":"Deep learning algorithms in the automatic segmentation of liver lesions in ultrasound investigations","volume":"12","author":"M\u0103muleanu","year":"2022","journal-title":"Life"},{"year":"2018","author":"Muneeswaran","key":"ref23"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"1744","DOI":"10.3390\/diagnostics13101744","article-title":"Detection of gallbladder disease types using deep learning: an informative medical method","volume":"13","author":"Obaid","year":"2023","journal-title":"Diagnostics"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"2413","DOI":"10.1002\/mp.14134","article-title":"Automatic prostate segmentation using deep learning on clinically diverse 3D transrectal ultrasound images","volume":"47","author":"Orlando","year":"2020","journal-title":"Med. Phys."},{"key":"ref26","doi-asserted-by":"publisher","first-page":"1548","DOI":"10.3390\/biom13101548","article-title":"Novel solution for using neural networks for kidney boundary extraction in 2d ultrasound data","volume":"13","author":"Peng","year":"2023","journal-title":"Biomol. Ther."},{"key":"ref27","doi-asserted-by":"publisher","first-page":"1177351","DOI":"10.3389\/fphys.2023.1177351","article-title":"Intelligent contour extraction approach for accurate segmentation of medical ultrasound images","volume":"14","author":"Peng","year":"2023","journal-title":"Front. Physiol."},{"year":"2017","author":"Ravishankar","key":"ref28"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"779","DOI":"10.1016\/j.bbe.2021.05.004","article-title":"A deep learning based approach for classification of abdominal organs using ultrasound images","volume":"41","author":"Reddy","year":"2021","journal-title":"Biocybern. Biomed. Eng."},{"key":"ref30","doi-asserted-by":"publisher","first-page":"309","DOI":"10.1145\/1015706.1015720","article-title":"Grabcut: interactive foreground extraction using iterated graph cuts","volume":"23","author":"Rother","year":"2004","journal-title":"ACM Trans. Graph."},{"key":"ref31","doi-asserted-by":"publisher","first-page":"8733","DOI":"10.1007\/s00330-021-07850-9","article-title":"Joint segmentation and classification of hepatic lesions in ultrasound images using deep learning","volume":"31","author":"Ryu","year":"2021","journal-title":"Eur. Radiol."},{"key":"ref32","doi-asserted-by":"publisher","first-page":"177","DOI":"10.14366\/usg.20085","article-title":"Current status of deep learning applications in abdominal ultrasonography","volume":"40","author":"Song","year":"2021","journal-title":"Ultrasonography"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"1932","DOI":"10.3892\/etm.2017.4209","article-title":"Abdominal ultrasonography for patients with abdominal pain as a first-line diagnostic imaging modality","volume":"13","author":"Tomizawa","year":"2017","journal-title":"Exp. Ther. Med."},{"key":"ref34","doi-asserted-by":"publisher","first-page":"1670","DOI":"10.1109\/TUFFC.2022.3161719","article-title":"Interpretable machine learning for characterization of focal liver lesions by contrast-enhanced ultrasound","volume":"69","author":"Turco","year":"2022","journal-title":"IEEE Trans. Ultrason. Ferroelectr. Freq. Control"},{"key":"ref35","doi-asserted-by":"publisher","first-page":"851644","DOI":"10.3389\/fmed.2022.851644","article-title":"Diabetic macular edema detection using end-to-end deep fusion model and anatomical landmark visualization on an edge computing device","volume":"9","author":"Wang","year":"2022","journal-title":"Front. Med."},{"key":"ref36","doi-asserted-by":"publisher","first-page":"534","DOI":"10.1007\/s00261-020-02564-w","article-title":"Deep learning for differentiation of benign and malignant solid liver lesions on ultrasonography","volume":"46","author":"Xi","year":"2021","journal-title":"Abdom. Radiol."},{"year":"2018","author":"Xu","key":"ref37"},{"year":"2017","author":"Xu","key":"ref38"},{"key":"ref39","doi-asserted-by":"publisher","first-page":"1034","DOI":"10.3390\/bioengineering11101034","article-title":"Advances in medical image segmentation: a comprehensive review of traditional, deep learning and hybrid approaches","volume":"11","author":"Xu","year":"2024","journal-title":"Bioengineering"},{"key":"ref40","doi-asserted-by":"publisher","first-page":"101602","DOI":"10.1016\/j.media.2019.101602","article-title":"Automatic kidney segmentation in ultrasound images using subsequent boundary distance regression and pixelwise classification networks","volume":"60","author":"Yin","year":"2020","journal-title":"Med. Image Anal."},{"year":"2019","author":"Yin","key":"ref41"},{"year":"2023","author":"Zhang","key":"ref42"},{"key":"ref43","doi-asserted-by":"publisher","first-page":"102338","DOI":"10.1016\/j.compmedimag.2024.102338","article-title":"SEG-LUS: a novel ultrasound segmentation method for liver and its accessory structures based on multi-head self-attention","volume":"113","author":"Zhang","year":"2024","journal-title":"Comput. Med. Imaging Graph."}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1618607\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,23]],"date-time":"2025-07-23T05:36:41Z","timestamp":1753249001000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1618607\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,23]]},"references-count":43,"alternative-id":["10.3389\/frai.2025.1618607"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1618607","relation":{},"ISSN":["2624-8212"],"issn-type":[{"type":"electronic","value":"2624-8212"}],"subject":[],"published":{"date-parts":[[2025,7,23]]},"article-number":"1618607"}}