{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T08:25:25Z","timestamp":1779265525232,"version":"3.51.4"},"reference-count":48,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,1,6]],"date-time":"2025-01-06T00:00:00Z","timestamp":1736121600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurorobot."],"abstract":"<jats:p>U-Net and its variants have been widely used in the field of image segmentation. In this paper, a lightweight multi-scale Ghost U-Net (MSGU-Net) network architecture is proposed. This can efficiently and quickly process image segmentation tasks while generating high-quality object masks for each object. The pyramid structure (SPP-Inception) module and ghost module are seamlessly integrated in a lightweight manner. Equipped with an efficient local attention (ELA) mechanism and an attention gate mechanism, they are designed to accurately identify the region of interest (ROI). The SPP-Inception module and ghost module work in tandem to effectively merge multi-scale information derived from low-level features, high-level features, and decoder masks at each stage. Comparative experiments were conducted between the proposed MSGU-Net and state-of-the-art networks on the ISIC2017 and ISIC2018 datasets. In short, compared to the baseline U-Net, our model achieves superior segmentation performance while reducing parameter and computation costs by 96.08 and 92.59%, respectively. Moreover, MSGU-Net can serve as a lightweight deep neural network suitable for deployment across a range of intelligent devices and mobile platforms, offering considerable potential for widespread adoption.<\/jats:p>","DOI":"10.3389\/fnbot.2024.1480055","type":"journal-article","created":{"date-parts":[[2025,1,6]],"date-time":"2025-01-06T07:02:53Z","timestamp":1736146973000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":3,"title":["MSGU-Net: a lightweight multi-scale ghost U-Net for image segmentation"],"prefix":"10.3389","volume":"18","author":[{"given":"Hua","family":"Cheng","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yang","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Huangxin","family":"Xu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dingliang","family":"Li","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zejian","family":"Zhong","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yinchuan","family":"Zhao","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhuo","family":"Yan","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2025,1,6]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"2481","DOI":"10.1109\/TPAMI.2016.2644615","article-title":"Segnet: a deep convolutional encoder-decoder architecture for image Seg-mentation","volume":"39","author":"Badrinarayanan","year":"2017","journal-title":"IEEE Trans. Pat. Anal Mac. Int."},{"key":"ref2","doi-asserted-by":"publisher","first-page":"633","DOI":"10.3390\/s23020633","article-title":"Segment-then-segment: context-preserving crop-based segmentation for large biomedical images","volume":"23","author":"Ben\u010devi\u0107","year":"2023","journal-title":"Sensors"},{"key":"ref3","volume-title":"ISIC 2017-skin lesion analysis towards melanoma detection","author":"Berseth","year":"2017"},{"key":"ref4","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-25066-8_9","article-title":"Swin-unet: Unet-like pure transformer for medical image segmentation","volume-title":"Computer vision \u2013 ECCV 2022 workshops. ECCV 2022. Lecture notes in computer science","author":"Cao","year":"2023"},{"key":"ref5","volume-title":"TransUNet: Transformers make strong En-coders for medical image segmentation","author":"Chen","year":"2021"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"357","DOI":"10.1080\/17476938708814211","article-title":"Semantic image segmentation with deep convolutional nets and fully connected CRFs","volume":"7","author":"Chen","year":"2014","journal-title":"Comput. Sci."},{"key":"ref7","doi-asserted-by":"publisher","first-page":"834","DOI":"10.1109\/TPAMI.2017.2699184","article-title":"DeepLab: semantic image segmentation with deep convolutional nets, Atrous convolution, and fully connected CRFs","volume":"40","author":"Chen","year":"2017","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref8","volume-title":"Rethinking Atrous convolution for semantic image segmentation","author":"Chen","year":"2017"},{"key":"ref9","first-page":"1251","volume-title":"Xception: Deep learning with Depthwise separable convolutions","author":"Chollet","year":"2017"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"424","DOI":"10.1007\/978-3-319-46723-8_49","article-title":"3D U-net: learning dense volumetric segmentation from sparse annotation","volume":"9901","author":"\u00c7i\u00e7ek","year":"2016","journal-title":"MICCAI."},{"key":"ref11","author":"Codella","year":"2019"},{"key":"ref12","author":"Dosovitskiy","year":"2021"},{"key":"ref13","volume-title":"RL-CoSeg: A novel image co-segmentation algorithm with deep reinforcement learning. Computer vision and pattern recognition","author":"Duan","year":"2022"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2203.00131","article-title":"A multi-scale transformer for medical image segmentation: architectures, model efficiency, and benchmark","author":"Gao","year":"2022"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"1580","DOI":"10.1109\/CVPR42600.2020.00165","article-title":"Ghostnet: more features from cheap operations","volume":"22","author":"Han","year":"2020","journal-title":"IEEE\/CVF CVPR."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"574","DOI":"10.1109\/WACV51458.2022.00181","article-title":"Unetr: transformers for 3d medical image segmentation","volume":"2022","author":"Hatamizadeh","year":"2022","journal-title":"IEEE\/CVF WCAV"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"315","DOI":"10.1093\/rasti\/rzae019","article-title":"Continunet: fast deep radio image segmentation in the square kilometre array era with u-net","volume":"3","author":"Hattie","year":"2024","journal-title":"RAS Techn."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"247388","DOI":"10.1371\/journal.pone.0247388","article-title":"SA-net: a scale-attention network for medical image segmentation","volume":"16","author":"Hu","year":"2021","journal-title":"PLoS One"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"1055","DOI":"10.1109\/ICASSP40776.2020.9053405","article-title":"Unet 3+: a full-scale connected Unet for medical image segmentation","volume":"4","author":"Huang","year":"2020","journal-title":"IEEE ICASSP."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"448","DOI":"10.48550\/arXiv.1502.03167","article-title":"Batch normalization: accelerating deep network training by reducing internal covariate shift","volume":"3167","author":"Ioffe","year":"2015","journal-title":"ICML."},{"key":"ref21","author":"Jensen","year":"2018"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"640","DOI":"10.1109\/CVPR.2015.7298965","article-title":"Fully convolutional networks for semantic segmentation","volume":"39","author":"Long","year":"2015","journal-title":"IEEE CVPR."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"893","DOI":"10.1007\/978-3-030-00934-299","article-title":"Y-net: joint segmentation and classification for Diagnosis of breast biopsy images","volume":"11","author":"Mehta","year":"2018","journal-title":"MICCAI."},{"key":"ref24","doi-asserted-by":"publisher","first-page":"565","DOI":"10.1109\/3DV.2016.79","article-title":"V-net: Fully convolutional neural networks for volumetric medical image segmentation","volume":"22","author":"Milletari","year":"2016","journal-title":"3DV"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"1874","DOI":"10.1007\/s10278-024-01070-5","article-title":"Gaunet: a lightweight ghost and attention u-net for medical image segmentation","volume":"37","author":"Pang","year":"2024","journal-title":"J. Imaging Inf. Med."},{"key":"ref26","doi-asserted-by":"publisher","first-page":"234","DOI":"10.1007\/978-3-319-24574-4_28","article-title":"U-net: convolutional networks for biomedical image segmentation","volume":"9351","author":"Ronneberger","year":"2015","journal-title":"MICCAI."},{"key":"ref27","doi-asserted-by":"publisher","first-page":"405","DOI":"10.1007\/978-3-031-43901-8_39","article-title":"MedNeXt: transformer-driven scaling of ConvNets for medical image segmentation. Medical image computing and computer assisted intervention","volume":"14223","author":"Saikat","year":"2023","journal-title":"MICCAI"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"1349","DOI":"10.3233\/jifs-221359","article-title":"An artificial bee colony-based framework for multi-objective optimization of three-way decisions with probabilistic rough sets","volume":"44","author":"Soumya","year":"2023","journal-title":"J. Int. Fuzzy Syst. Appl. Eng. Technol."},{"key":"ref29","doi-asserted-by":"publisher","first-page":"11231","DOI":"10.1609\/aaai.v31i1.11231","article-title":"Inception-v4, inception-ResNet and the impact of residual connections on learning","volume":"31","author":"Szegedy","year":"2017","journal-title":"AAAI Conf. Art. Int."},{"key":"ref30","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/CVPR.2015.7298594","article-title":"Going deeper with convolutions","volume":"2015","author":"Szegedy","year":"2015","journal-title":"IEEE CVPR."},{"key":"ref31","author":"Szegedy","year":"2016"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"9969","DOI":"10.48550\/arXiv.2211.12905","article-title":"GhostNetv2: enhancing cheap operation with Long-range attention","volume":"35","author":"Tang","year":"2022","journal-title":"36th Conf. Neural Inf. Proc. Syst. (NeurIPS 2022)"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/978-3-030-87193-2_4","article-title":"Medical transformer: gated axial-attention for medical image Segmentation","volume":"12901","author":"Valanarasu","year":"","journal-title":"MICCAI."},{"key":"ref34","doi-asserted-by":"publisher","first-page":"23","DOI":"10.1007\/978-3-031-16443-9_3","article-title":"Unext: Mlp-based rapid medical image segmentation network","volume":"13435","author":"Valanarasu","year":"2022","journal-title":"MICCAI."},{"key":"ref35","doi-asserted-by":"publisher","first-page":"363","DOI":"10.1007\/978-3-030-59719-1_36","article-title":"M. Kiu-net: towards accurate segmentation of biomedical images using over-complete representations","volume":"12264","author":"Valanarasu","year":"2020","journal-title":"MICCAI. Part IV"},{"key":"ref36","doi-asserted-by":"publisher","first-page":"965","DOI":"10.1109\/TMI.2021.3130469","article-title":"Kiu-net: Overcomplete convolutional architectures for BioMedical image and volumetric segmentation","volume":"41","author":"Valanarasu","year":"","journal-title":"IEEE Trans. Med. Imaging."},{"key":"ref37","doi-asserted-by":"publisher","first-page":"109","DOI":"10.1007\/978-3-030-87193-2_11","article-title":"TransBTS: multimodal brain tumor segmentation using transformer. Medical image computing and computer assisted intervention (MICCAI 2021)","author":"Wang","year":"2021","journal-title":"Comput. Sci."},{"key":"ref38","doi-asserted-by":"publisher","first-page":"102327","DOI":"10.1016\/j.media.2021.102327","article-title":"FAT-net: feature adaptive transformers for automated skin lesion segmentation","volume":"76","author":"Wu","year":"2022","journal-title":"Med. Img. Anal."},{"key":"ref39","doi-asserted-by":"publisher","first-page":"6744","DOI":"10.1109\/TSMC.2021.3096065","article-title":"A latent factor analysis-based approach to online sparse streaming feature selection","volume":"52","author":"Wu","year":"2022","journal-title":"IEEE Trans. Syst. Man Cybernetics Syst."},{"key":"ref40","doi-asserted-by":"publisher","first-page":"575","DOI":"10.1109\/TSC.2023.3331570","article-title":"MMLF: multi-metric latent feature analysis for high-dimensional and incomplete data","volume":"17","author":"Wu","year":"2024","journal-title":"IEEE Trans. Serv. Comput."},{"key":"ref41","doi-asserted-by":"publisher","first-page":"6608","DOI":"10.1109\/TMI.2024.3436608","article-title":"Self-supervised medical image segmentation using deep reinforced adaptive masking","volume":"12","author":"Xu","year":"","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref42","doi-asserted-by":"publisher","first-page":"1857","DOI":"10.1109\/TNSE.2023.3332810","article-title":"Collaborative attention guided multi-scale feature fusion network for medical image segmentation","volume":"11","author":"Xu","year":"","journal-title":"IEEE Trans Netw Sci Eng"},{"key":"ref43","volume-title":"ELA: efficient local attention for deep convolutional neural networks","author":"Xu","year":"2024"},{"key":"ref44","doi-asserted-by":"publisher","first-page":"4976","DOI":"10.1109\/TFUZZ.2024.3399769","article-title":"Fuzziness-based three-way decision with neighborhood rough sets under the framework of shadowed sets","volume":"32","author":"Yang","year":"2024","journal-title":"IEEE Trans. Fuzzy Syst."},{"key":"ref45","doi-asserted-by":"publisher","first-page":"158","DOI":"10.3390\/info11030158","article-title":"A multi-task framework for action prediction","volume":"11","author":"Yu","year":"2020","journal-title":"Information."},{"key":"ref46","doi-asserted-by":"publisher","first-page":"14","DOI":"10.1007\/978-3-030-87193-2_2","article-title":"Transfuse: fusing transformers and CNNs for medical image segmentation","volume":"12901","author":"Zhang","year":"2021","journal-title":"MICCAI."},{"key":"ref47","doi-asserted-by":"publisher","first-page":"102656","DOI":"10.1016\/j.media.2022.102656","article-title":"Multi-modal contrastive mutual learning and pseudo-label re-learning for semi-supervised medical image segmentation","volume":"83","author":"Zhang","year":"2023","journal-title":"Med. Image Anal."},{"key":"ref48","doi-asserted-by":"publisher","first-page":"1856","DOI":"10.1109\/TMI.2019.2959609","article-title":"Unet++: redesigning skip connections to exploit multiscale features in image segmentation","volume":"39","author":"Zhou","year":"2019","journal-title":"IEEE Trans. Med. Imging."}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2024.1480055\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,1,6]],"date-time":"2025-01-06T07:02:59Z","timestamp":1736146979000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2024.1480055\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,1,6]]},"references-count":48,"alternative-id":["10.3389\/fnbot.2024.1480055"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2024.1480055","relation":{},"ISSN":["1662-5218"],"issn-type":[{"value":"1662-5218","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,1,6]]},"article-number":"1480055"}}