{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T14:19:25Z","timestamp":1778681965796,"version":"3.51.4"},"reference-count":29,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T00:00:00Z","timestamp":1778630400000},"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. Comput. Neurosci."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>The process of precise structural evaluation for paranasal sinuses based on CT scan data establishes a foundation for medical professionals to assess human anatomical variations, supporting the diagnosis and treatment of ear, nose, and throat (ENT) conditions. Existing deep learning methods face difficulties in analyzing complex sinus structures due to limited annotated datasets and lack of clinical interpretability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This study presents an explainable hybrid CNN-Transformer framework incorporating a self-supervised 3D convolutional autoencoder to perform structural analysis of paranasal sinus CT volumes. The framework is evaluated on the multi-institutional CT-SCOPE dataset, which contains diverse scan data from different hospitals and CT scanner models. The proposed approach combines anatomical segmentation to generate precise boundaries with residual-based structural representation learning for anomaly detection without requiring pathology labels.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The hybrid segmentation model achieves high anatomical fidelity, producing Dice similarity coefficients above 0.83 across all four sinus regions, including maxillary, ethmoid, frontal, and sphenoid sinuses. The architecture integrates convolutional feature extraction with Transformer-based contextual modeling to capture both fine structural details and global anatomical context. The self-supervised autoencoder generates reconstruction residual maps that highlight structural deviations from standard osseous patterns. Grad-CAM-based interpretability analysis demonstrates that the model focuses on clinically relevant regions such as sinus walls and ethmoidal partitions.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>The integration of accurate segmentation, residual-based structural representation learning, and explainable attention mechanisms provides a comprehensive framework for anatomical analysis of paranasal sinus CT images. This approach establishes a foundation for future pathology-aware clinical modeling and supports the development of reliable AI-assisted diagnostic systems.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fncom.2026.1778260","type":"journal-article","created":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T13:50:05Z","timestamp":1778680205000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Explainable hybrid CNN-transformer with self-supervised learning for structural analysis of paranasal sinus CT"],"prefix":"10.3389","volume":"20","author":[{"given":"Najeeb","family":"Ullah","sequence":"first","affiliation":[{"name":"College of Computer and Systems Engineering, Abdullah Al Salem University","place":["Khaldiya, Kuwait"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shabbab Ali","family":"Algamdi","sequence":"additional","affiliation":[{"name":"Department of Software Engineering","place":["Prince Sattam bin Abdulaziz University, Saudi Arabia"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tariq","family":"Sadad","sequence":"additional","affiliation":[{"name":"Department of Computer Science, University of Engineering and Technology","place":["Mardan, Pakistan"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,5,13]]},"reference":[{"key":"B1","first-page":"429","article-title":"Supervised contrastive learning to classify paranasal anomalies in the maxillary sinus","volume-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention","author":"Bhattacharya","year":"2022"},{"key":"B2","doi-asserted-by":"publisher","first-page":"32","DOI":"10.21037\/ajo-22-9","article-title":"Sinus mucosal thickening on magnetic resonance imaging and inflammation.","volume":"5","author":"Chang","year":"2022","journal-title":"Aust. J. Otolaryngol."},{"key":"B3","article-title":"A simple framework for contrastive learning of visual representations","volume-title":"Proceedings of the 37th International Conference on Machine Learning","author":"Chen","year":"2020"},{"key":"B4","article-title":"Bernardo Avila Pires. Bootstrap your own latent: A new approach to self-supervised learning","volume-title":"Proceedings of the 34th International Conference on Neural Information Processing Systems","author":"Grill","year":"2020"},{"key":"B5","doi-asserted-by":"publisher","first-page":"976","DOI":"10.1007\/s00330-002-1652-3","article-title":"A new low-dose CT examination compared with standard-dose CT in the diagnosis of acute sinusitis.","volume":"13","author":"Hagtvedt","year":"2003","journal-title":"Eur. Radiol."},{"key":"B6","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR52688.2022.01553","article-title":"Masked autoencoders are scalable vision learners","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"He","year":"2022"},{"key":"B7","doi-asserted-by":"publisher","first-page":"165","DOI":"10.4103\/smj.smj_59_20","article-title":"Incidental sinus findings in brain MRI for suspected intracranial disease.","volume":"24","author":"Ikubor","year":"2021","journal-title":"Sahel Med. J."},{"key":"B8","first-page":"2789","article-title":"Automatic segmentation of the paranasal sinus from computer tomography images using a probabilistic atlas and a fully convolutional network","volume-title":"Proceedings of the 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)","author":"Iwamoto","year":"2019"},{"key":"B9","doi-asserted-by":"publisher","first-page":"250","DOI":"10.3390\/diagnostics11020250","article-title":"Deep learning for diagnosis of paranasal sinusitis using multi-view radiographs.","volume":"11","author":"Jeon","year":"2021","journal-title":"Diagnostics"},{"key":"B10","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":"B11","doi-asserted-by":"publisher","first-page":"e0263125","DOI":"10.1371\/journal.pone.0263125","article-title":"Detection of maxillary sinus fungal ball via 3-D CNN-based artificial intelligence.","volume":"17","author":"Kim","year":"2022","journal-title":"PLoS One"},{"key":"B12","doi-asserted-by":"publisher","first-page":"7","DOI":"10.1097\/RLI.0000000000000503","article-title":"Deep learning in diagnosis of maxillary sinusitis using conventional radiography.","volume":"49","author":"Kim","year":"2019","journal-title":"Imaging Sci. Dent."},{"key":"B13","doi-asserted-by":"publisher","first-page":"718","DOI":"10.3390\/tomography8020059","article-title":"Semi-supervised deep learning semantic segmentation for 3D CT scoring of chronic rhinosinusitis.","volume":"8","author":"Kuo","year":"2022","journal-title":"Tomography"},{"key":"B14","doi-asserted-by":"publisher","first-page":"1189","DOI":"10.3390\/math10071189","article-title":"Fully automatic segmentation, identification and preoperative planning for nasal surgery of sinuses.","volume":"10","author":"Kuo","year":"2022","journal-title":"Mathematics"},{"key":"B15","doi-asserted-by":"publisher","first-page":"32","DOI":"10.1002\/rcs.175","article-title":"Image-enhanced surgical navigation for endoscopic sinus surgery.","volume":"4","author":"Lapeer","year":"2008","journal-title":"Int. J. Med. Robot. Comput. Assist. Surg."},{"key":"B16","doi-asserted-by":"publisher","first-page":"3675","DOI":"10.3390\/app11083675","article-title":"Image-guided endoscopic sinus surgery with 3D volumetric visualization.","volume":"11","author":"Lee","year":"2021","journal-title":"Appl. Sci."},{"key":"B17","doi-asserted-by":"publisher","first-page":"1052","DOI":"10.1002\/alr.22014","article-title":"Three-dimensional image analysis for staging chronic rhinosinusitis.","volume":"7","author":"Lim","year":"2017","journal-title":"Int. Forum Allergy Rhinol."},{"key":"B18","doi-asserted-by":"publisher","first-page":"621","DOI":"10.1002\/lio2.436","article-title":"Augmented reality-enhanced navigation in endoscopic sinus surgery.","volume":"5","author":"Linxweiler","year":"2020","journal-title":"Laryngoscope Investig. Otolaryngol."},{"key":"B19","doi-asserted-by":"publisher","first-page":"1025","DOI":"10.1002\/alr.22958","article-title":"Deep learning classification of inverted papilloma malignant transformation.","volume":"12","author":"Liu","year":"2022","journal-title":"Int. Forum Allergy Rhinol."},{"key":"B20","doi-asserted-by":"publisher","first-page":"709","DOI":"10.5812\/ircmj.1797","article-title":"Assessment of the diagnostic accuracy of limited CT scan of paranasal sinuses in the identification of sinusitis.","volume":"14","author":"Noorian","year":"2012","journal-title":"Iran. Red Crescent Med. J."},{"key":"B21","doi-asserted-by":"publisher","first-page":"963","DOI":"10.1002\/alr.21219","article-title":"Three-dimensional volumetric computed tomographic scoring for chronic rhinosinusitis.","volume":"3","author":"Pallanch","year":"2013","journal-title":"Int. Forum Allergy Rhinol."},{"key":"B22","doi-asserted-by":"publisher","first-page":"e13459","DOI":"10.7759\/cureus.12727","article-title":"Anatomical variations of the nasal cavities and paranasal sinuses: A systematic review.","volume":"13","author":"Papadopoulou","year":"2021","journal-title":"Cureus"},{"key":"B23","doi-asserted-by":"publisher","first-page":"507","DOI":"10.1007\/s00405-008-0777-7","article-title":"CT-based manual segmentation and evaluation of paranasal sinuses.","volume":"266","author":"Pirner","year":"2009","journal-title":"Eur. Arch. Otorhinolaryngol."},{"key":"B24","doi-asserted-by":"publisher","first-page":"846","DOI":"10.1002\/alr.21552","article-title":"Volumetric CT analysis of the olfactory cleft in chronic patients with rhinosinusitis.","volume":"5","author":"Soler","year":"2015","journal-title":"Int. Forum Allergy Rhinol."},{"key":"B25","doi-asserted-by":"publisher","first-page":"32087","DOI":"10.1038\/s41598-025-17266-w","article-title":"Comparison of segmentation performance of CNNs, vision transformers, and hybrid networks for paranasal sinuses.","volume":"15","author":"Song","year":"2025","journal-title":"Sci. Rep."},{"key":"B26","doi-asserted-by":"publisher","first-page":"111962","DOI":"10.1016\/j.dib.2025.111962","article-title":"CT-SCOPE: annotated dataset of CT scans for the automatic semantic segmentation of the osseous structures of the paranasal sinuses","volume":"62","author":"Sun","year":"2025","journal-title":"Data Brief"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.15085103","article-title":"CT-SCOPE: Annotated dataset of CT scans for automatic semantic segmentation of paranasal sinuses.","author":"Sun","year":"2025","journal-title":"Zenodo"},{"key":"B28","doi-asserted-by":"publisher","first-page":"2","DOI":"10.1097\/00000441-199807000-00002","article-title":"Functional anatomy and computed tomography imaging of the paranasal sinuses.","volume":"316","author":"Taxy","year":"1998","journal-title":"Am. J. Med. Sci."},{"key":"B29","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-32251-9_42","article-title":"Models genesis: Generic autodidactic models for 3D medical image analysis","volume-title":"Proceedings of the First International Conference on Medical Image Computing and Computer-Assisted Intervention","author":"Zhou","year":"2019"}],"container-title":["Frontiers in Computational Neuroscience"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fncom.2026.1778260\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T13:50:08Z","timestamp":1778680208000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fncom.2026.1778260\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,13]]},"references-count":29,"alternative-id":["10.3389\/fncom.2026.1778260"],"URL":"https:\/\/doi.org\/10.3389\/fncom.2026.1778260","relation":{},"ISSN":["1662-5188"],"issn-type":[{"value":"1662-5188","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,13]]},"article-number":"1778260"}}