{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T05:51:38Z","timestamp":1791525098163,"version":"4.3.4"},"reference-count":61,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T00:00:00Z","timestamp":1791504000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001839","name":"University Grants Committee","doi-asserted-by":"publisher","award":["UGC\/FDS24\/E20\/25"],"award-info":[{"award-number":["UGC\/FDS24\/E20\/25"]}],"id":[{"id":"10.13039\/501100001839","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001839","name":"University Grants Committee","doi-asserted-by":"publisher","award":["GRF1421622"],"award-info":[{"award-number":["GRF1421622"]}],"id":[{"id":"10.13039\/501100001839","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001839","name":"University Grants Committee","doi-asserted-by":"publisher","award":["ITS\/240\/21"],"award-info":[{"award-number":["ITS\/240\/21"]}],"id":[{"id":"10.13039\/501100001839","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001839","name":"University Grants Committee","doi-asserted-by":"publisher","award":["GDX20220530111005039"],"award-info":[{"award-number":["GDX20220530111005039"]}],"id":[{"id":"10.13039\/501100001839","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Digit. Health"],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Manual segmentation and evaluation of optical coherence tomography (OCT) images are difficult and rely on specialist training, which limits the output in a busy clinic. While automatic segmentation can ease this burden, the deep models that currently lead the field tend to be too heavy for routine deployment on standard clinical hardware. A further obstacle is the unusual geometry of OCT data: instead of compact objects on a background, the structures of interest form long, undulating tissue strata that extend horizontally across the whole frame, and this layout is poorly matched to convolutional kernels designed for ordinary photographs.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We address both issues with LiDeOCTNet, a compact segmentation backbone tailored to the layered nature of OCT. Within a reduced UNet, we introduce learnable sampling offsets so that each kernel can reshape itself locally and trace the curved, irregular interfaces between tissues. To assess the design, we evaluate the network on three OCT collections covering retinal as well as endoscopic acquisitions.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Across all three, LiDeOCTNet surpasses a parameter-matched UNet baseline and approaches the accuracy of much heavier benchmarks such as TransUNet, despite using about 750\u00d7 fewer parameters and running roughly 7.4\u00d7 faster at inference.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>Together these properties point to LiDeOCTNet as a practical option for real-time OCT analysis pipelines, including embedded and resource-limited devices.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fdgth.2026.1892323","type":"journal-article","created":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T05:29:51Z","timestamp":1791523791000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A robust lightweight deep learning network with deformable convolutions for efficient segmentation of optical coherence tomography images"],"prefix":"10.3389","volume":"8","author":[{"given":"Mehmood","family":"Nawaz","sequence":"first","affiliation":[{"name":"Department of Data Science and Digital Innovation, The Hong Kong Shue Yan University","place":["Hong Kong SAR, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Syed Muhammad Tariq","family":"Abbasi","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, The Chinese University of Hong Kong","place":["Hong Kong SAR, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sheheryar","family":"Khan","sequence":"additional","affiliation":[{"name":"Division of Science, Engineering, and Health Studies (SEHS), School of Professional Education and Executive Development, The Hong Kong Polytechnic University","place":["Hong Kong SAR, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Muhammad Ameer","family":"Hamza","sequence":"additional","affiliation":[{"name":"Department of Data Science and Digital Innovation, The Hong Kong Shue Yan University","place":["Hong Kong SAR, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Khadija","family":"Bibi","sequence":"additional","affiliation":[{"name":"Department of Science, English Schools Foundation College, Sha Tin","place":["Hong Kong SAR, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Syeda Aimen","family":"Abbasi","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, The Chinese University of Hong Kong","place":["Hong Kong SAR, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Muhammad","family":"Daud","sequence":"additional","affiliation":[{"name":"Department of Agriculture, Bahauddin Zakariya University","place":["Multan, Pakistan"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Muhammad","family":"Asim","sequence":"additional","affiliation":[{"name":"Division of Science, Engineering, and Health Studies (SEHS), School of Professional Education and Executive Development, The Hong Kong Polytechnic University","place":["Hong Kong SAR, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wu","family":"Yuan","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, The Chinese University of Hong Kong","place":["Hong Kong SAR, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,10,9]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1126","DOI":"10.1364\/BOE.10.001126","article-title":"Parallel deep neural networks for endoscopic OCT image segmentation","volume":"10","author":"Li","year":"2019","journal-title":"Biomed Opt Express"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1413","DOI":"10.1109\/TMI.2022.3228285","article-title":"Retifluidnet: a self-adaptive and multi-attention deep convolutional network for retinal OCT fluid segmentation","volume":"42","author":"Rasti","year":"2022","journal-title":"IEEE Trans Med Imaging"},{"key":"B3","doi-asserted-by":"publisher","first-page":"2405","DOI":"10.1364\/BOE.8.002405","article-title":"Endoscopic optical coherence tomography: technologies and clinical applications","volume":"8","author":"Gora","year":"2017","journal-title":"Biomed Opt Express"},{"key":"B4","doi-asserted-by":"publisher","first-page":"57","DOI":"10.3390\/diagnostics4020057","article-title":"Endoscopic optical coherence tomography for clinical gastroenterology","volume":"4","author":"Tsai","year":"2014","journal-title":"Diagnostics"},{"key":"B5","doi-asserted-by":"publisher","first-page":"2091","DOI":"10.1364\/BOE.453396","article-title":"In vivo assessment of inflammatory bowel disease in rats with ultrahigh-resolution colonoscopic OCT","volume":"13","author":"Yuan","year":"2022","journal-title":"Biomed Opt Express"},{"key":"B6","doi-asserted-by":"publisher","first-page":"238","DOI":"10.1109\/TBME.2022.3188173","article-title":"Direct visualization and quantitative imaging of small airway anatomy in vivo using deep learning assisted diffractive OCT","volume":"70","author":"Yuan","year":"2023","journal-title":"IEEE Trans Biomed Eng"},{"key":"B7","doi-asserted-by":"publisher","first-page":"375","DOI":"10.1080\/00051144.2021.1973298","article-title":"Annotated retinal optical coherence tomography images (AROI) database for joint retinal layer and fluid segmentation","volume":"62","author":"Melin\u0161\u010dak","year":"2021","journal-title":"Automatika"},{"key":"B8","doi-asserted-by":"publisher","first-page":"6326","DOI":"10.1364\/BOE.434775","article-title":"Connectivity-based deep learning approach for segmentation of the epithelium in in vivo human esophageal OCT images","volume":"12","author":"Yang","year":"2021","journal-title":"Biomed Opt Express"},{"key":"B9","doi-asserted-by":"publisher","first-page":"303","DOI":"10.1109\/TIP.2019.2931461","article-title":"Deep neural network regression for automated retinal layer segmentation in optical coherence tomography images","volume":"29","author":"Ngo","year":"2019","journal-title":"IEEE Trans Image Process"},{"key":"B10","first-page":"1451","article-title":"Understanding convolution for semantic segmentation","author":"Wang","year":"2018"},{"key":"B11","doi-asserted-by":"publisher","first-page":"2697","DOI":"10.1364\/BOE.8.002697","article-title":"Automatic and robust segmentation of endoscopic OCT images and optical staining","volume":"8","author":"Zhang","year":"2017","journal-title":"Biomed Opt Express"},{"key":"B12","first-page":"3431","article-title":"Fully convolutional networks for semantic segmentation","author":"Long","year":"2015"},{"key":"B13","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-04061-1","volume-title":"Computer Aided Intervention and Diagnostics in Clinical and Medical Images","author":"Peter","year":"2019"},{"key":"B14","doi-asserted-by":"publisher","first-page":"e70141","DOI":"10.1002\/nbm.70141","article-title":"Artificial intelligence\u2013based approaches for brain tumor segmentation in MRI: a review","volume":"38","author":"Bibi","year":"2025","journal-title":"NMR Biomed"},{"key":"B15","doi-asserted-by":"publisher","first-page":"3051","DOI":"10.1109\/TMI.2025.3556870","article-title":"Semi-supervised knee cartilage segmentation with successive eigen noise-assisted mean teacher knowledge distillation","volume":"44","author":"Khan","year":"2025","journal-title":"IEEE Trans Med Imaging"},{"key":"B16","first-page":"1610","article-title":"Accuracy improvement of semantic segmentation using appropriate datasets for robot navigation","author":"Miyamoto","year":"2019"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1257","DOI":"10.1007\/s00521-017-3158-6","article-title":"Medical image semantic segmentation based on deep learning","volume":"29","author":"Jiang","year":"2018","journal-title":"Neural Comput Appl"},{"key":"B18","doi-asserted-by":"publisher","first-page":"83002","DOI":"10.1109\/ACCESS.2021.3086530","article-title":"Deep neural architectures for medical image semantic segmentation","volume":"9","author":"Khan","year":"2021","journal-title":"IEEE Access"},{"key":"B19","first-page":"800","article-title":"Convolutional neural network based automatic plaque characterization for intracoronary optical coherence tomography images","author":"He","year":"2018"},{"key":"B20","doi-asserted-by":"publisher","first-page":"748","DOI":"10.1002\/lsm.23414","article-title":"Deep learning in biomedical optics","volume":"53","author":"Tian","year":"2021","journal-title":"Lasers Surg Med"},{"key":"B21","doi-asserted-by":"publisher","first-page":"626","DOI":"10.1016\/j.neucom.2022.08.021","article-title":"Deep learning in retinal optical coherence tomography (OCT): a comprehensive survey","volume":"493","author":"Viedma","year":"2022","journal-title":"Neurocomputing"},{"key":"B22","doi-asserted-by":"publisher","first-page":"626","DOI":"10.1016\/j.neucom.2022.01.005","article-title":"Review the state-of-the-art technologies of semantic segmentation based on deep learning","volume":"493","author":"Mo","year":"2022","journal-title":"Neurocomputing"},{"key":"B23","first-page":"59","article-title":"Optical coherence tomography (OCT): principle and technical realization","author":"Aumann","year":"2019"},{"key":"B24","first-page":"3213","article-title":"The cityscapes dataset for semantic urban scene understanding","author":"Cordts","year":"2016"},{"key":"B25","first-page":"248","article-title":"Imagenet: a large-scale hierarchical image database","author":"Deng","year":"2009"},{"key":"B26","author":"Pont-Tuset","year":"2017"},{"key":"B27","first-page":"3234","article-title":"The SYNTHIA dataset: a large collection of synthetic images for semantic segmentation of urban scenes","author":"Ros","year":"2016"},{"key":"B28","doi-asserted-by":"publisher","first-page":"4481","DOI":"10.1364\/BOE.9.004481","article-title":"Robust layer segmentation of esophageal OCT images based on graph search using edge-enhanced weights","volume":"9","author":"Gan","year":"2018","journal-title":"Biomedl Optics Express"},{"key":"B29","doi-asserted-by":"publisher","first-page":"354","DOI":"10.1016\/j.patcog.2017.10.013","article-title":"Recent advances in convolutional neural networks","volume":"77","author":"Gu","year":"2018","journal-title":"Pattern Recognit"},{"key":"B30","doi-asserted-by":"publisher","first-page":"2281","DOI":"10.1109\/TMI.2019.2903562","article-title":"CE-net: context encoder network for 2D medical image segmentation","volume":"38","author":"Gu","year":"2019","journal-title":"IEEE Trans Med Imaging"},{"key":"B31","author":"Yu","year":"2015"},{"key":"B32","first-page":"7794","article-title":"Non-local neural networks","author":"Wang","year":"2018"},{"key":"B33","first-page":"801","article-title":"Encoder-decoder with atrous separable convolution for semantic image segmentation","author":"Chen","year":"2018"},{"key":"B34","author":"Dosovitskiy","year":"2020"},{"key":"B35","author":"Chen","year":"2021"},{"key":"B36","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3505244","article-title":"Transformers in vision: a survey","volume":"54","author":"Khan","year":"2022","journal-title":"ACM Comput Surv"},{"key":"B37","doi-asserted-by":"publisher","first-page":"576","DOI":"10.1016\/j.neucom.2020.07.143","article-title":"Automatic fluid segmentation in retinal optical coherence tomography images using attention based deep learning","volume":"452","author":"Liu","year":"2021","journal-title":"Neurocomputing"},{"key":"B38","author":"Mehta","year":"2021"},{"key":"B39","first-page":"10347","article-title":"Training data-efficient image transformers distillation through attention","author":"Touvron","year":"2021"},{"key":"B40","first-page":"1296","article-title":"Swiftnet: real-time video object segmentation","author":"Wang","year":"2021"},{"key":"B41","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TIM.2024.3400305","article-title":"Lightweight retinal layer segmentation with global reasoning","volume":"73","author":"He","year":"2024","journal-title":"IEEE Trans Instrum Meas"},{"key":"B42","doi-asserted-by":"publisher","first-page":"60","DOI":"10.3390\/e27010060","article-title":"An efficient retinal fluid segmentation network based on large receptive field context capture for optical coherence tomography images","volume":"27","author":"Qi","year":"2025","journal-title":"Entropy"},{"key":"B43","doi-asserted-by":"publisher","first-page":"207","DOI":"10.3390\/biomimetics10040207","article-title":"FRNet V2: a lightweight full-resolution convolutional neural network for OCTA vessel segmentation","volume":"10","author":"Gao","year":"2025","journal-title":"Biomimetics"},{"key":"B44","doi-asserted-by":"publisher","first-page":"101856","DOI":"10.1016\/j.media.2020.101856","article-title":"Structured layer surface segmentation for retina OCT using fully convolutional regression networks","volume":"68","author":"He","year":"2021","journal-title":"Med Image Anal"},{"key":"B45","doi-asserted-by":"publisher","first-page":"102269","DOI":"10.1016\/j.compmedimag.2023.102269","article-title":"Unravelling the complexity of optical coherence tomography image segmentation using machine and deep learning techniques: a review","volume":"108","author":"Nawaz","year":"2023","journal-title":"Comput Med Imaging Graph"},{"key":"B46","first-page":"764","article-title":"Deformable convolutional networks","author":"Dai","year":"2017"},{"key":"B47","doi-asserted-by":"publisher","first-page":"149","DOI":"10.1016\/j.knosys.2019.04.025","article-title":"DUNet: a deformable network for retinal vessel segmentation","volume":"178","author":"Jin","year":"2019","journal-title":"Knowl Based Syst"},{"key":"B48","doi-asserted-by":"publisher","first-page":"68","DOI":"10.1109\/TRPMS.2021.3059780","article-title":"DefED-net: deformable encoder-decoder network for liver and liver tumor segmentation","volume":"6","author":"Lei","year":"2021","journal-title":"IEEE Trans Radiat Plasma Med Sci"},{"key":"B49","doi-asserted-by":"publisher","first-page":"43669","DOI":"10.1109\/ACCESS.2022.3168693","article-title":"Blazeneo: blazing fast polyp segmentation and neoplasm detection","volume":"10","author":"An","year":"2022","journal-title":"IEEE Access"},{"key":"B50","first-page":"3552","article-title":"HarDNet: a low memory traffic network","author":"Chao","year":"2019"},{"key":"B51","author":"Iandola","year":"2014"},{"key":"B52","first-page":"234","article-title":"U-net: convolutional networks for biomedical image segmentation","author":"Ronneberger","year":"2015"},{"key":"B53","first-page":"448","article-title":"Batch normalization: accelerating deep network training by reducing internal covariate shift","volume-title":"Proceedings of Machine Learning Research (PMLR)","author":"Ioffe","year":"2015"},{"key":"B54","author":"Agarap","year":"2018"},{"key":"B55","doi-asserted-by":"publisher","first-page":"e202000042","DOI":"10.1002\/jbio.202000042","article-title":"Pseudo-real-time retinal layer segmentation for high-resolution adaptive optics optical coherence tomography","volume":"13","author":"Janpongsri","year":"2020","journal-title":"J Biophotonics"},{"key":"B56","doi-asserted-by":"publisher","first-page":"18922","DOI":"10.1109\/JSEN.2022.3193718","article-title":"Hand gestures classification using electrical impedance tomography images","volume":"22","author":"Nawaz","year":"2022","journal-title":"|3294"},{"key":"B57","doi-asserted-by":"publisher","first-page":"7173","DOI":"10.1007\/s12652-021-03570-6","article-title":"Object detection and segmentation by composition of fast fuzzy C-mean clustering based maps","volume":"14","author":"Nawaz","year":"2023","journal-title":"J Ambient Intell Humaniz Comput"},{"key":"B58","first-page":"116","article-title":"Shufflenet V2: practical guidelines for efficient CNN architecture design","author":"Ma","year":"2018"},{"key":"B59","first-page":"4510","article-title":"Mobilenetv2: inverted residuals and linear bottlenecks","author":"Sandler","year":"2018"},{"key":"B60","first-page":"568","article-title":"Pyramid vision transformer: a versatile backbone for dense prediction without convolutions","author":"Wang","year":"2021"},{"key":"B61","doi-asserted-by":"publisher","first-page":"102035","DOI":"10.1016\/j.media.2021.102035","article-title":"Loss odyssey in medical image segmentation","volume":"71","author":"Ma","year":"2021","journal-title":"Med Image Anal"}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1892323\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T05:29:52Z","timestamp":1791523792000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1892323\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,9]]},"references-count":61,"alternative-id":["10.3389\/fdgth.2026.1892323"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1892323","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,9]]},"article-number":"1892323"}}