{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T06:27:10Z","timestamp":1787120830442,"version":"build-2736575974"},"reference-count":32,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T00:00:00Z","timestamp":1787097600000},"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. Artif. Intell."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>In laparoscopic surgery, which is a minimally invasive procedure, endoscopic cameras are very common in visual guidance. Nevertheless, surgical smoke produced during electrocautery and laser ablation greatly impairs the quality of images produced by decreasing contrast, blurring edges, and distorting color data. This deterioration is experienced with both the visibility of the surgeon and computer vision-based surgical assistance system performance. Physics-based and image enhancement models are ineffective due to unrealistic assumptions of the smoke distribution, whereas deep learning methods are typically complicated.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>It is based on the Progressive Frequency-Aware Network (PFAN), which is a lightweight CNN-Transformer hybrid that isolates smoke and tissue signal in the frequency domain. PFAN uses Multi-scale Bottleneck-Inverting (MBI) blocks to restore high-frequency details and Locally-Enhanced Axial Attention Transformer (LAT) blocks to learn global contextual details. In order to improve the quality of reconstruction and structural preservation, we suggest another Hybrid Attention-Residual UNet model that incorporates residual learning and dual attention (channel and spatial attention) to a conditional GAN. To enhance the elimination and localization of smoke compared to PFAN, the suggested hybrid model aims at spatial refinement of features and adaptive weighting of features.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The experimental findings prove that the suggested hybrid model can be competitive in terms of PSNR and SSIM, but with the preservation of structural consistency and enhanced visual clarity.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>Frequency-aware learning with attention-based refinement is a solid solution to laparoscopic image de-smoking.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/frai.2026.1898106","type":"journal-article","created":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T05:44:38Z","timestamp":1787118278000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A lightweight hybrid CNN-transformer model for smoke removal in laparoscopic surgery images"],"prefix":"10.3389","volume":"9","author":[{"given":"Dhinesh Kumar","family":"Mahalingam","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tamilarasi Kathirvel","family":"Murugan","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Logeswari","family":"Govindaraj","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sanjay","family":"Chauhan","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Vishal","family":"Babu","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,19]]},"reference":[{"key":"ref2","author":"Chen","year":"2022"},{"key":"ref9001","doi-asserted-by":"publisher","first-page":"1615","DOI":"10.1109\/TMI.2019.2953717","article-title":"De-smokeGCN: Generative Cooperative Networks for Joint Surgical Smoke Detection and Removal","volume":"39","author":"Chen","year":"2019","journal-title":"IEEE Transactions on Medical Imaging."},{"key":"ref1","author":"Chen","year":"2026"},{"key":"ref9002","doi-asserted-by":"publisher","first-page":"6773","DOI":"10.1109\/TIP.2020.2993407","article-title":"PMHLD: Patch map-based hybrid learning DehazeNet for single image haze removal","volume":"29","author":"Chen","year":"2020","journal-title":"IEEE Transactions on Image Processing"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"372","DOI":"10.1038\/s43588-026-00973-3","article-title":"HorusEye: a self-supervised foundation model for generalizable X-ray tomography restoration","volume":"6","author":"Chu","year":"2026","journal-title":"Nat Comput Sci"},{"key":"ref4","author":"Fu","year":"2016"},{"key":"ref5","volume-title":"Endoscopic Image super-resolution Algorithm using edge and Disparity Awareness","author":"Hayat","year":"2023"},{"key":"ref001","doi-asserted-by":"publisher","first-page":"5802","DOI":"10.1109\/CVPR52688.2022.00572","article-title":"Image dehazing transformer with transmission-aware 3D position embedding","author":"Guo","year":"2022","journal-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"ref6","author":"Hu","year":"2021"},{"key":"ref7","author":"Li","year":"2017"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"1236","DOI":"10.3390\/horticulturae11101236","article-title":"Lightweight CNN\u2013transformer hybrid network with contrastive learning for few-shot noxious weed recognition","volume":"11","author":"Li","year":"2025","journal-title":"Horticulturae"},{"key":"ref11","first-page":"102447","article-title":"DeSmoke-LAP: unpaired laparoscopic smoke removal via domain-specific losses","volume":"79","author":"Liu","year":"2022","journal-title":"Med. Image Anal."},{"key":"ref10","author":"Liu","year":"2019"},{"key":"ref9","first-page":"102693","article-title":"L-SAHGNet: lightness-aware hybrid guided network for real-time laparoscopic smoke removal","volume":"84","author":"Liu","year":"2023","journal-title":"Med. Image Anal."},{"key":"ref9009","doi-asserted-by":"publisher","first-page":"102691","DOI":"10.1007\/s11042-026-21347-x","article-title":"Attention-aware network with lightness embedding and Hybrid Guided Embedding for laparoscopic image desmoking","volume":"127","author":"Liu","year":"2026","journal-title":"Computerized Medical Imaging and Graphics"},{"key":"ref9010","doi-asserted-by":"publisher","first-page":"885","DOI":"10.1007\/s11548-022-02595-2","article-title":"DeSmoke-LAP: Improved unpaired image-to-image translation for desmoking in laparoscopic surgery","volume":"17","author":"Pan","year":"2022","journal-title":"International Journal of Computer Assisted Radiology and Surgery"},{"key":"ref12","author":"Qin","year":"2020"},{"key":"ref9011","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2511.05059","author":"Sheng","year":"2025","journal-title":"SurgiATM: A Physics-Guided Plug-and-Play Model for Deep Learning-Based Smoke Removal in Laparoscopic Surgery"},{"key":"ref13","first-page":"10432","article-title":"DS-cGAN: surgical smoke removal and depth estimation for robotic surgery","volume":"7","author":"Shen","year":"2022","journal-title":"IEEE Robot. Autom. Lett."},{"key":"ref14","first-page":"10891","article-title":"Transformer-based image restoration: a survey","volume":"45","author":"Song","year":"2023","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref15","author":"Song","year":"2023"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"139","DOI":"10.1186\/s12938-018-0590-5","article-title":"Variational based smoke removal in laparoscopic images","volume":"17","author":"Wang","year":"2018","journal-title":"Biomed. Eng. Online"},{"key":"ref9003","doi-asserted-by":"publisher","first-page":"505","DOI":"10.1117\/12.2507822","article-title":"Multiscale deep desmoking for laparoscopic surgery. In Medical Imaging 2019: Image Processing","volume":"10949","author":"Wang","year":"2019","journal-title":"SPIE."},{"key":"ref9004","first-page":"4534","article-title":"PMHLD: Patch Map Based Hybrid Learning DehazeNet,","volume":"30","author":"Wang","year":"2021","journal-title":"IEEE Transactions on Image Processing"},{"key":"ref18","first-page":"2871","article-title":"PFAN: progressive frequency-aware network for lightweight laparoscopic image desmoking","volume":"33","author":"Wang","year":"2023","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"ref9005","first-page":"307","article-title":"Self-supervised video desmoking for laparoscopic surgery","volume-title":"European Conference on Computer Vision","author":"Wu","year":"2024"},{"key":"ref9006","doi-asserted-by":"publisher","first-page":"165","DOI":"10.1109\/ACCESS.2015.2511558","article-title":"Review of video and image defogging algorithms and related studies on image restoration and enhancement","volume":"4","author":"Xu","year":"2016","journal-title":"IEEE Access."},{"key":"ref20","author":"Zamir","year":"2022"},{"key":"ref9007","doi-asserted-by":"publisher","first-page":"6547","DOI":"10.1109\/TVCG.2023.3347438","article-title":"A Self-Supervised Network-Based Smoke Removal and Depth Estimation for Monocular Endoscopic Videos","volume":"30","author":"Zhang","year":"2024","journal-title":"IEEE Transactions on Visualization and Computer Graphics"},{"key":"ref23","first-page":"2101","article-title":"Residual Swin transformer network for surgical smoke removal","volume":"41","author":"Zhang","year":"2022","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"670","DOI":"10.1109\/TCBB.2022.3204673","article-title":"Synchronizing detection and removal of smoke in endoscopic images with cyclic consistency adversarial nets","volume":"21","author":"Zhou","year":"2022","journal-title":"IEEE\/ACM Trans. Comput. Biol. Bioinform."},{"key":"ref26","author":"Zhu","year":"2018"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1898106\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T05:44:40Z","timestamp":1787118280000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1898106\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,19]]},"references-count":32,"alternative-id":["10.3389\/frai.2026.1898106"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1898106","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,19]]},"article-number":"1898106"}}