Skip to content
2000
Volume 21, Issue 1
  • ISSN: 1573-4056
  • E-ISSN: 1875-6603
file format text download EPUB
file format pdf download PDF
side by side viewer icon HTML

Abstract

Background

The sphenoid sinus is an important access point for trans-sphenoidal surgeries, but variations in its pneumatization may complicate surgical safety. Deep learning can be used to identify these anatomical variations.

Methods

We developed a convolutional neural network (CNN) model for the automated prediction of sphenoid sinus pneumatization patterns in computed tomography (CT) scans. This model was tested on mid-sagittal CT images. Two radiologists labeled all CT images into four pneumatization patterns: Conchal (type I), presellar (type II), sellar (type III), and postsellar (type IV). We then augmented the training set to address the limited size and imbalanced nature of the data.

Results

The initial dataset included 249 CT images, divided into training ( = 174) and test ( = 75) datasets. The training dataset was augmented to 378 images. Following augmentation, the overall diagnostic accuracy of the model improved from 76.71% to 84%, with an area under the curve (AUC) of 0.84, indicating very good diagnostic performance. Subgroup analysis showed excellent results for type IV, with the highest AUC of 0.93, perfect sensitivity (100%), and an F1-score of 0.94. The model also performed robustly for type I, achieving an accuracy of 97.33% and high specificity (99%). These metrics highlight the model's potential for reliable clinical application.

Conclusion

The proposed CNN model demonstrates very good diagnostic accuracy in identifying various sphenoid sinus pneumatization patterns, particularly excelling in type IV, which is crucial for endoscopic sinus surgery due to its higher risk of surgical complications. By assisting radiologists and surgeons, this model enhances the safety of transsphenoidal surgery, highlighting its value, novelty, and applicability in clinical settings.

2025 © 2025 The Author(s). Published by Bentham Science Publisher.
This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0), a copy of which is available at: https://creativecommons.org/licenses/by/4.0/legalcode. This license permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Loading

Article metrics loading...

/content/journals/cmir/10.2174/0115734056363158250429101521
2025-05-22
2025-10-30

  • From This Site

    /content/journals/cmir/10.2174/0115734056363158250429101521
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

/deliver/fulltext/cmir/21/1/CMIR-21-E15734056363158.html?itemId=/content/journals/cmir/10.2174/0115734056363158250429101521&mimeType=html&fmt=ahah

References

  1. VaeziA. CardenasE. Pinheiro-NetoC. PaluzziA. BranstetterB.F.IV GardnerP.A. SnydermanC.H. Fernandez-MirandaJ.C. Classification of sphenoid sinus pneumatization: relevance for endoscopic skull base surgery.Laryngoscope2015125357758110.1002/lary.2498925417777
     [Google Scholar]
  2. AçarG. GökşanA.S. AydoğduD. Computed tomography based evaluation of the association between sphenoid sinus pneumatization patterns and variations of adjacent bony structures in relation to age and gender.Neurosurg. Rev.202447134910.1007/s10143‑024‑02594‑839046640
     [Google Scholar]
  3. ŠtokovićN. TrkuljaV. Dumić-ČuleI. Čuković-BagićI. LaucT. VukičevićS. Sphenoid sinus types, dimensions and relationship with surrounding structures.Annals. Anatomy.20162036976
     [Google Scholar]
  4. SagarS. JahanS. KashyapS.K. Prevalence of anatomical variations of sphenoid sinus and its adjacent structures pneumatization and its significance: A CT scan study.Indian J. Otolaryngol. Head Neck Surg.20237542979298910.1007/s12070‑023‑03879‑y37974780
     [Google Scholar]
  5. TesfayeS. HambaN. GerbiA. NegeriZ. Radio-anatomic variability in sphenoid sinus pneumatization with its relationship to adjacent anatomical structures and their impact upon reduction of complications following endonasal transsphenoidal surgeries.Translat. Res. Anatomy20212410012610.1016/j.tria.2021.100126
     [Google Scholar]
  6. Nasi-KordhishtiI. GieseS. HirtB. HoneggerJ. Anatomy of the pituitary region.Pituitary Tumors. HoneggerJ. ReinckeM. PetersennS. Academic Press2021678510.1016/B978‑0‑12‑819949‑7.00061‑5
     [Google Scholar]
  7. CellinaM. GibelliD. FloridiC. ToluianT. Valenti PittinoC. MartinenghiC. OlivaG. Sphenoid sinuses: Pneumatisation and anatomical variants—what the radiologist needs to know and report to avoid intraoperative complications.Surg. Radiol. Anat.20204291013102410.1007/s00276‑020‑02490‑y32394118
     [Google Scholar]
  8. AijazA. AhmedH. FahmiS. SamreenT. RasheedB. JabeenH. A comprehensive computed tomographic analysis of pneumatization pattern of sphenoid sinus and their association with protrusion/dehiscence of vital neurovascular structures in a pakistani subgroup.Turk Neurosurg.202333350150810.5137/1019‑5149.JTN.40154‑22.336951035
     [Google Scholar]
  9. LawsE.R.Jr Vascular complications of transsphenoidal surgery.Pituitary19992216317010.1023/A:100995191764911081167
     [Google Scholar]
  10. Seredyka-BurdukM. BurdukP.K. WierzchowskaM. KaluznyB. MalukiewiczG. Ophthalmic complications of endoscopic sinus surgery.Rev. Bras. Otorrinolaringol.201783331832327233691
     [Google Scholar]
  11. HamidO. El FikyL. HassanO. KotbA. El FikyS. Anatomic variations of the sphenoid sinus and their impact on trans-sphenoid pituitary surgery.Off. J. North Am. Skull Base Soc.200818191510.1055/s‑2007‑992764
     [Google Scholar]
  12. RasemanJ. GuryildirimM. Beer-FurlanA. JhaveriM. TajudeenB.A. ByrneR.W. BatraP.S. Preoperative computed tomography imaging of the sphenoid sinus: Striving towards safe transsphenoidal surgery.J. Neurol. Surg. B Skull Base202081325126210.1055/s‑0039‑169183132499999
     [Google Scholar]
  13. OsieG. Darbari KaulR. AlvaradoR. KatsoulotosG. RimmerJ. KalishL. CampbellR.G. SacksR. HarveyR.J. A scoping review of artificial intelligence research in rhinology.Am. J. Rhinol. Allergy202337443844810.1177/1945892423116243736895144
     [Google Scholar]
  14. LeonardiR. Lo GiudiceA. FarronatoM. RonsivalleV. AllegriniS. MusumeciG. Fully automatic segmentation of sinonasal cavity and pharyngeal airway based on convolutional neural networks.American J. Orthodont. Dentofacial Orthoped.2021159682483510.1016/j.ajodo.2020.05.017
     [Google Scholar]
  15. ParmarP. HabibA-R. MendisD. DanielA. DuvnjakM. HoJ. SmithM. RoshanD. WongE. SinghN. An artificial intelligence algorithm that identifies middle turbinate pneumatisation (Concha bullosa) on sinus computed tomography scans.J. Laryngol. Otol.2020134432833110.1017/S002221512000044432234081
     [Google Scholar]
  16. HuangJ. HabibA-R. MendisD. ChongJ. SmithM. DuvnjakM. ChiuC. SinghN. WongE. An artificial intelligence algorithm that differentiates anterior ethmoidal artery location on sinus computed tomography scans.J. Laryngol. Otol.20201341525510.1017/S002221511900253631865928
     [Google Scholar]
  17. LauraC.O. HofmannP. DrechslerK. WesargS. Automatic detection of the nasal cavities and paranasal sinuses using deep neural networks.2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019)08-11 April 2019, Venice, Italy, 2019, pp. 1154-1157.10.1109/ISBI.2019.8759481
     [Google Scholar]
  18. XuJ. WangS. ZhouZ. LiuJ. JiangX. ChenX. Automatic CT image segmentation of maxillary sinus based on VGG network and improved V-Net.Int. J. CARS20201591457146510.1007/s11548‑020‑02228‑632676871
     [Google Scholar]
  19. ShettyS. MubarakA.S. R DavidL. Al JouhariM.O. TalaatW. Al-RawiN. AlKawasS. ShettyS. Uzun OzsahinD. The application of mask region-based convolutional neural networks in the detection of nasal septal deviation using cone beam computed tomography images: Proof-of-concept study.JMIR Form. Res.20248e5733510.2196/5733539226096
     [Google Scholar]
  20. LimS.H. KimJ.H. KimY.J. ChoM.Y. JungJ.U. HaR. Aux-MVNet: Auxiliary classifier-based multi-view convolutional neural network for maxillary sinusitis diagnosis on paranasal sinuses view.Diagnostics202212373610.3390/diagnostics12030736
     [Google Scholar]
  21. KuoC.F.J. LiaoY.S. BarmanJ. LiuS.C. Semi-supervised deep learning semantic segmentation for 3D volumetric computed tomographic scoring of chronic rhinosinusitis: Clinical correlations and comparison with lund-mackay scoring.Tomography20228271872910.3390/tomography802005935314636
     [Google Scholar]
  22. LiuG.S. YangA. KimD. HojelA. VoevodskyD. WangJ. TongC.C.L. UngererH. PalmerJ.N. KohanskiM.A. NayakJ.V. HwangP.H. AdappaN.D. PatelZ.M. Deep learning classification of inverted papilloma malignant transformation using 3D convolutional neural networks and magnetic resonance imaging.Int. Forum Allergy Rhinol.20221281025103310.1002/alr.2295834989484
     [Google Scholar]
  23. NakagawaJ. FujimaN. HirataK. TangM. TsunetaS. SuzukiJ. HaradaT. IkebeY. HommaA. KanoS. MinowaK. KudoK. Utility of the deep learning technique for the diagnosis of orbital invasion on CT in patients with a nasal or sinonasal tumor.Cancer Imaging20222215210.1186/s40644‑022‑00492‑036138422
     [Google Scholar]
  24. SeolY.J. KimY.J. KimY.S. CheonY.W. KimK.G. A study on 3D deep learning-based automatic diagnosis of nasal fractures.Sensors (Basel)202222250610.3390/s2202050635062465
     [Google Scholar]
  25. LiY. LiuX. ZhuangX. WangM. SongX. Assessment of low-dose paranasal sinus CT imaging using a new deep learning image reconstruction technique in children compared to adaptive statistical iterative reconstruction V (ASiR-V).BMC Med. Imaging202222110610.1186/s12880‑022‑00834‑135658908
     [Google Scholar]
  26. KimD.K. ChoB.J. LeeM.J. KimJ.H. Prediction of age and sex from paranasal sinus images using a deep learning network.Medicine (Baltimore)20211007e2475610.1097/MD.000000000002475633607821
     [Google Scholar]
  27. TaylorA. HabibA.R. KumarA. WongE. HasanZ. SinghN. An artificial intelligence algorithm for the classification of sphenoid sinus pneumatisation on sinus computed tomography scans.Clinical Otolaryngol.202348688889410.1111/coa.14088
     [Google Scholar]
  28. HambergerC.A. HammerG. NorlenG. SjogrenB. Transantrosphenoidal hypophysectomy.Arch. otolaryngol.1961742810.1001/archotol.1961.00740030005002
     [Google Scholar]
  29. GüldnerC. PistoriusS.M. DiogoI. BienS. SesterhennA. WernerJ.A. Analysis of pneumatization and neurovascular structures of the sphenoid sinus using cone-beam tomography (CBT).Acta radiologica201253221421910.1258/ar.2011.110381
     [Google Scholar]
  30. LeCunY. BengioY. HintonG. Deep learning.Nature2015521755343644410.1038/nature1453926017442
     [Google Scholar]
  31. RazaR. ZulfiqarF. KhanM.O. ArifM. AlviA. IftikharM.A. AlamT. Lung-EffNet: Lung cancer classification using EfficientNet from CT-scan images.Eng. Appl. Artif. Intell.202312610690210.1016/j.engappai.2023.106902
     [Google Scholar]
  32. ZhangH. QieY. Applying deep learning to medical imaging: A review.Appl. Sci. (Basel)202313181052110.3390/app131810521
     [Google Scholar]
  33. Smith-BindmanR. MigliorettiD.L. LarsonE.B. Rising use of diagnostic medical imaging in a large integrated health system.Health Aff. (Millwood)20082761491150210.1377/hlthaff.27.6.149118997204
     [Google Scholar]
  34. KrupinskiE.A. BerbaumK.S. CaldwellR.T. SchartzK.M. KimJ. Long radiology workdays reduce detection and accommodation accuracy.J. Am. Coll. Radiol.20107969870410.1016/j.jacr.2010.03.00420816631
     [Google Scholar]
  35. YasakaK. AbeO. Deep learning and artificial intelligence in radiology: Current applications and future directions.PLoS Med.20181511e100270710.1371/journal.pmed.100270730500815
     [Google Scholar]
  36. TadavarthiY. MakeevaV. WagstaffW. ZhanH. PodlasekA. BhatiaN. HeilbrunM. KrupinskiE. SafdarN. BanerjeeI. GichoyaJ. TrivediH. Overview of noninterpretive artificial intelligence models for safety, quality, workflow, and education applications in radiology practice.Radiol. Artif. Intell.202242e21011410.1148/ryai.21011435391770
     [Google Scholar]
  37. MuksimovaS. UmirzakovaS. MardievaS. ChoY.I. Enhancing medical image denoising with innovative teacher–student model-based approaches for precision diagnostics.Sensors (Basel)20232323950210.3390/s2323950238067873
     [Google Scholar]
  38. WigginsW.F. MagudiaK. SchmidtT.M.S. O’ConnorS.D. CarrC.D. KohliM.D. AndrioleK.P. Imaging AI in practice: A demonstration of future workflow using integration standards.Radiol. Artif. Intell.202136e21015210.1148/ryai.202121015234870224
     [Google Scholar]
  39. HuangM. ZhangX.S. BhattiU.A. WuY. ZhangY. Yasin GhadiY. An interpretable approach using hybrid graph networks and explainable AI for intelligent diagnosis recommendations in chronic disease care.Biomed. Signal Process. Control20249110591310.1016/j.bspc.2023.105913
     [Google Scholar]
  40. FamurewaO.C. IbitoyeB.O. AmeyeS.A. AsaleyeC.M. AyoolaO.O. OnigbindeO.S. Sphenoid sinus pneumatization, septation, and the internal carotid artery: A computed tomography study.Nigerian Med. J.2018591713
     [Google Scholar]
  41. FaddaG.L. PetrelliA. UrbanelliA. CastelnuovoP. BignamiM. CrosettiE. SuccoG. CavalloG. Risky anatomical variations of sphenoid sinus and surrounding structures in endoscopic sinus surgery.Head Face Med.20221812910.1186/s13005‑022‑00336‑z36057720
     [Google Scholar]
  42. DoganM.E. KotanlıS. YavuzY. WahjuningrumD.A. PawarA.M. Computed tomography-based assessment of sphenoid sinus and sella turcica pneumatization analysis: A retrospective study.PeerJ202311e1662310.7717/peerj.1662338130934
     [Google Scholar]
  43. ChengP.M. MontagnonE. YamashitaR. PanI. Cadrin-ChênevertA. Perdigón RomeroF. ChartrandG. KadouryS. TangA. Deep learning: An update for radiologists.Radiographics20214151427144510.1148/rg.202120021034469211
     [Google Scholar]
  44. SalmanS. LiuX. Overfitting mechanism and avoidance in deep neural networks.arXiv:1901065662019
     [Google Scholar]
  45. AlzubaidiL. ZhangJ. HumaidiA.J. Al-DujailiA. DuanY. Al-ShammaO. SantamaríaJ. FadhelM.A. Al-AmidieM. FarhanL. Review of deep learning: Concepts, CNN architectures, challenges, applications, future directions.J. Big Data2021815310.1186/s40537‑021‑00444‑833816053
     [Google Scholar]
  46. ShortenC. KhoshgoftaarT.M. A survey on image data augmentation for deep learning.J. Big Data2019616010.1186/s40537‑019‑0197‑0
     [Google Scholar]
  47. MohamedT.I.A. OyeladeO.N. EzugwuA.E. Automatic detection and classification of lung cancer CT scans based on deep learning and ebola optimization search algorithm.PLoS One2023188e028579610.1371/journal.pone.028579637590282
     [Google Scholar]
  48. HanH. WangW.Y. MaoB.H. Borderline-SMOTE: A new over-sampling method in imbalanced data sets learning.Advances in Intelligent Computing. ICIC 2005. Lecture Notes in Computer ScienceBerlin, HeidelbergSpringer200510.1007/11538059_91
     [Google Scholar]
  49. FotouhiS. AsadiS. KattanM.W. A comprehensive data level analysis for cancer diagnosis on imbalanced data.J. Biomed. Inform.20199010308910.1016/j.jbi.2018.12.00330611011
     [Google Scholar]
  50. Werner de VargasV. Schneider ArandaJ.A. dos Santos CostaR. da Silva PereiraP.R. Victória BarbosaJ.L. Imbalanced data preprocessing techniques for machine learning: A systematic mapping study.Knowl. Inf. Syst.2023651315710.1007/s10115‑022‑01772‑836405957
     [Google Scholar]
  51. KhushiM. ShaukatK. AlamT.M. HameedI.A. UddinS. LuoS. YangX. ReyesM.C. A comparative performance analysis of data resampling methods on imbalance medical data.IEEE Access2021910996010997510.1109/ACCESS.2021.3102399
     [Google Scholar]
  52. MoraisM. CalistoF.M. SantiagoC. AleluiaC. NascimentoJ.C. Classification of breast cancer in mri with multimodal fusion.2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI)18-21 April 2023, Cartagena, Colombia, 2023, pp. 1-4.
     [Google Scholar]
  53. BeddiarD.R. OussalahM. MuhammadU. SeppänenT. A Deep learning based data augmentation method to improve COVID-19 detection from medical imaging.Knowl. Base. Syst.202328011098510.1016/j.knosys.2023.110985
     [Google Scholar]
  54. MuksimovaS. UmirzakovaS. KangS. ChoY.I. CerviLearnNet: Advancing cervical cancer diagnosis with reinforcement learning-enhanced convolutional networks.Heliyon2024109e2991310.1016/j.heliyon.2024.e2991338694035
     [Google Scholar]
  55. HiremathS.B. GautamA.B. SheejaK. BenjaminG. Assessment of variations in sphenoid sinus pneumatization in Indian population: A multidetector computed tomography study.Indian J. Radiol. Imaging201828327327910.4103/ijri.IJRI_70_1830319202
     [Google Scholar]
  56. RahmatiA. GhafariR. AnjomShoaM. Normal variations of sphenoid sinus and the adjacent structures detected in cone beam computed tomography.J. Dent. (Shiraz)2016171323726966706
     [Google Scholar]
  57. KangY.J. ChoJ.H. KimD.H. KimS.W. Relationships of sphenoid sinus pneumatization with internal carotid artery characteristics.PLoS One2022178e027354510.1371/journal.pone.027354536006952
     [Google Scholar]
  58. ScuderiA.J. HarnsbergerH.R. BoyerR.S. Pneumatization of the paranasal sinuses: Normal features of importance to the accurate interpretation of CT scans and MR images.AJR Am. J. Roentgenol.199316051101110410.2214/ajr.160.5.84705858470585
     [Google Scholar]
  59. FujiokaM. YoungL.W. The sphenoidal sinuses: Radiographic patterns of normal development and abnormal findings in infants and children.Radiology1978129113313610.1148/129.1.133693864
     [Google Scholar]
  60. WhangboJ. LeeJ. KimY.J. KimS.T. KimK.G. Deep learning-based multi-class segmentation of the paranasal sinuses of sinusitis patients based on computed tomographic images.Sensors (Basel)2024246193310.3390/s2406193338544195
     [Google Scholar]
  61. MorganN. Van GervenA. SmoldersA. de Faria VasconcelosK. WillemsH. JacobsR. Convolutional neural network for automatic maxillary sinus segmentation on cone-beam computed tomographic images.Sci. Rep.2022121752310.1038/s41598‑022‑11483‑335525857
     [Google Scholar]
  62. JungS.K. LimH.K. LeeS. ChoY. SongI.S. Deep active learning for automatic segmentation of maxillary sinus lesions using a convolutional neural network.Diagnostics202111468810.3390/diagnostics11040688
     [Google Scholar]
  63. BhattiU.A. HuangM. Neira-MolinaH. MarjanS. BaryalaiM. TangH. WuG. BazaiS.U. MFFCG – Multi feature fusion for hyperspectral image classification using graph attention network.Expert Syst. Appl.202322912049610.1016/j.eswa.2023.120496
     [Google Scholar]
  64. NizamaniA.H. ChenZ. NizamaniA.A. BhattiU.A. Advance brain tumor segmentation using feature fusion methods with deep U-Net model with CNN for MRI data.J. King Saud Uni. - Comp. Inform. Sci.202335910179310.1016/j.jksuci.2023.101793
     [Google Scholar]
  65. HumphriesS.M. CentenoJ.P. NotaryA.M. GerowJ. CicchettiG. KatialR.K. BeswickD.M. RamakrishnanV.R. AlamR. LynchD.A. Volumetric assessment of paranasal sinus opacification on computed tomography can be automated using a convolutional neural network.Int. Forum Allergy Rhinol.202010111218122510.1002/alr.2258832306522
     [Google Scholar]
  66. MasseyC.J. RamosL. BeswickD.M. RamakrishnanV.R. HumphriesS.M. Clinical validation and extension of an automated, deep learning–based algorithm for quantitative sinus CT Analysis.AJNR Am. J. Neuroradiol.20224391318132410.3174/ajnr.A761636538385
     [Google Scholar]
  67. YanX. LinB. FuJ. LiS. WangH. FanW. FanY. FengM. WangR. FanJ. QiS. JiangC. Deep-learning-based automatic segmentation and classification for craniopharyngiomas.Front. Oncol.202313104884110.3389/fonc.2023.104884137213305
     [Google Scholar]
  68. FengT. FangY. PeiZ. LiZ. ChenH. HouP. WeiL. WangR. WangS. A convolutional neural network model for detecting sellar floor destruction of pituitary adenoma on magnetic resonance imaging scans.Front. Neurosci.20221690051910.3389/fnins.2022.90051935860294
     [Google Scholar]
  69. NicolaesJ. RaeymaeckersS. RobbenD. WilmsG. VandermeulenD. LibanatiC. Detection of vertebral fractures in CT using 3D convolutional neural networks.Computational Methods and Clinical Applications for Spine ImagingShenzhen, ChinaSpringer2019
     [Google Scholar]
  70. TomitaN. CheungY.Y. HassanpourS. Deep neural networks for automatic detection of osteoporotic vertebral fractures on CT scans.Comput. Biol. Med.20189881510.1016/j.compbiomed.2018.05.01129758455
     [Google Scholar]
  71. BhattiU.A. TangH. WuG. MarjanS. HussainA. Deep learning with graph convolutional networks: An overview and latest applications in computational intelligence.Int. J. Intell. Syst.202320231834210410.1155/2023/8342104
     [Google Scholar]
  72. SongX. LiH. GaoW. ChenY. WangT. MaG. LeiB. Augmented multicenter graph convolutional network for COVID-19 diagnosis.IEEE Trans. Industr. Inform.20211796499650910.1109/TII.2021.305668637981914
     [Google Scholar]
  73. HuangM. ZouJ. ZhangY. BhattiU.A. ChenJ. Efficient click-based interactive segmentation for medical image with improved Plain-ViT.IEEE J. Biomed. Health Inform.2024PP11210.1109/JBHI.2024.339289338656851
     [Google Scholar]
/content/journals/cmir/10.2174/0115734056363158250429101521
Loading
Deep Learning for Automated Prediction of Sphenoid Sinus Pneumatization in Computed Tomography
Curr. Med. Imaging 21, E15734056363158 (2025); https://doi.org/10.2174/0115734056363158250429101521
/content/journals/cmir/10.2174/0115734056363158250429101521
/content/journals/cmir/10.2174/0115734056363158250429101521
Loading

Data & Media loading...


  • Article Type:     Research Article
Keyword(s): Artificial intelligence; Computed tomography; Convolutional neural networks; Deep learning; Paranasal sinuses; Pneumatization; Sinonasal imaging; Sphenoid sinus

Most Cited Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error
Please enter a valid_number test