Automated segmentation and deep learning classification of ductopenic parotid salivary glands in sialo cone-beam CT images.

Classification Deep Learning Ductopenia Parotid salivary gland imaging Segmentation Sialo cone-beam CT scans

Journal

International journal of computer assisted radiology and surgery
ISSN: 1861-6429
Titre abrégé: Int J Comput Assist Radiol Surg
Pays: Germany
ID NLM: 101499225

Informations de publication

Date de publication:
31 Jul 2024
Historique:
received: 18 02 2024
accepted: 17 07 2024
medline: 1 8 2024
pubmed: 1 8 2024
entrez: 31 7 2024
Statut: aheadofprint

Résumé

This study addressed the challenge of detecting and classifying the severity of ductopenia in parotid glands, a structural abnormality characterized by a reduced number of salivary ducts, previously shown to be associated with salivary gland impairment. The aim of the study was to develop an automatic algorithm designed to improve diagnostic accuracy and efficiency in analyzing ductopenic parotid glands using sialo cone-beam CT (sialo-CBCT) images. We developed an end-to-end automatic pipeline consisting of three main steps: (1) region of interest (ROI) computation, (2) parotid gland segmentation using the Frangi filter, and (3) ductopenia case classification with a residual neural network (RNN) augmented by multidirectional maximum intensity projection (MIP) images. To explore the impact of the first two steps, the RNN was trained on three datasets: (1) original MIP images, (2) MIP images with predefined ROIs, and (3) MIP images after segmentation. Evaluation was conducted on 126 parotid sialo-CBCT scans of normal, moderate, and severe ductopenic cases, yielding a high performance of 100% for the ROI computation and 89% for the gland segmentation. Improvements in accuracy and F1 score were noted among the original MIP images (accuracy: 0.73, F1 score: 0.53), ROI-predefined images (accuracy: 0.78, F1 score: 0.56), and segmented images (accuracy: 0.95, F1 score: 0.90). Notably, ductopenic detection sensitivity was 0.99 in the segmented dataset, highlighting the capabilities of the algorithm in detecting ductopenic cases. Our method, which combines classical image processing and deep learning techniques, offers a promising solution for automatic detection of parotid glands ductopenia in sialo-CBCT scans. This may be used for further research aimed at understanding the role of presence and severity of ductopenia in salivary gland dysfunction.

Identifiants

pubmed: 39085681
doi: 10.1007/s11548-024-03240-w
pii: 10.1007/s11548-024-03240-w
doi:

Types de publication

Journal Article

Langues

eng

Sous-ensembles de citation

IM

Subventions

Organisme : Ministry of Science, Technology and Space
ID : 5007

Informations de copyright

© 2024. CARS.

Références

Ship JA (2002) Diagnosing, managing, and preventing salivary gland disorders. Oral Dis 8(2):77–89. https://doi.org/10.1034/j.1601-0825.2002.2o837.x
doi: 10.1034/j.1601-0825.2002.2o837.x pubmed: 11991308
Benito DA, Badger C, Hoffman HT, Joshi A (2020) Recommended imaging for salivary gland disorders. Curr Otorhinolaryngol Rep 8:311–320. https://doi.org/10.1007/s40136-020-00299-2
doi: 10.1007/s40136-020-00299-2
Abdalla-Aslan R, Keshet N, Zadik Y, Aframian DJ, Nadler C (2021) Standardization of terminology, imaging features, and interpretation of CBCT sialography of major salivary glands: a clinical review. Quintessence Int 52(8):728–740. https://doi.org/10.3290/j.qi.b1492217
doi: 10.3290/j.qi.b1492217 pubmed: 34076380
Abdel-Wahed N, Amer ME, Abo-Taleb NSM (2013) Assessment of the role of cone beam computed sialography in diagnosing salivary gland lesions. Imaging Sci Dent 43(1):17–23. https://doi.org/10.5624/isd.2013.43.1.17
doi: 10.5624/isd.2013.43.1.17 pubmed: 23524990 pmcid: 3604366
Jadu FM, Lam EWN (2013) A comparative study of the diagnostic capabilities of 2D plain radiograph and 3D cone beam CT sialography. Dentomaxillofac Radiol 42(1):20110319. https://doi.org/10.1259/dmfr.20110319
doi: 10.1259/dmfr.20110319 pubmed: 23253564 pmcid: 5083116
Keshet N, Aricha A, Friedlander-Barenboim S, Aframian DJ, Nadler C (2019) Novel parotid sialo-cone-beam computerized tomography features in patients with suspected Sjogren’s syndrome. Oral Dis 25(1):126–132. https://doi.org/10.1111/odi.12946
doi: 10.1111/odi.12946 pubmed: 30070035
Morosi C, Civelli E, Battiston C, Schiavo M, Mazzaferro V, Severini A, Marchianò A (2009) CT cholangiography: assessment of feasibility and diagnostic reliability. Eur J Radiol 72(1):114–117. https://doi.org/10.1016/j.ejrad.2008.05.011
doi: 10.1016/j.ejrad.2008.05.011 pubmed: 18602784
Salas-Silva S, Simoni-Nieves A, Lopez-Ramirez J, Bucio L, Gómez-Quiroz LE, Gutiérrez-Ruiz MC, Roma MG (2019) Cholangiocyte death in ductopenic cholestatic cholangiopathies: Mechanistic basis and emerging therapeutic strategies. Life Sci 218:324–339. https://doi.org/10.1016/j.lfs.2018.12.044
doi: 10.1016/j.lfs.2018.12.044 pubmed: 30610870
Amiel T, Shauly O, Istoyler EG, Alterman M, Keshet N, Mazor S, Pikovsky A, Aframian DJ, Joskowicz L, Nadler C (2024) The anatomical pattern of ductal arborization in parotid glands using cone-beam computerized sialography. Clin Anat. https://doi.org/10.1002/ca.24136
doi: 10.1002/ca.24136 pubmed: 39044623
Shauly O, Joskowicz L, Istoyler EG, Nadler C (2021) Parotid salivary ductal system segmentation and modeling in sialo-CBCT scans. Comput Methods Biomech Biomed Eng Imaging Vis 9(5):488–499. https://doi.org/10.1080/21681163.2020.1866670
doi: 10.1080/21681163.2020.1866670
Heo MS, Kim JE, Hwang JJ, Han SS, Kim JS, Yi WJ, Park IW (2021) Artificial intelligence in oral and maxillofacial radiology: What is currently possible? Dentomaxillofac Radiol 50(3):20200375. https://doi.org/10.1259/dmfr.20200375
doi: 10.1259/dmfr.20200375 pubmed: 33197209
Chandni SM, Kushwaha AKS (2023) The power of deep learning for intelligent tumor classification systems: a review. Comput Electr Eng 106:108586. https://doi.org/10.1016/j.compeleceng.2023.108586
doi: 10.1016/j.compeleceng.2023.108586
Mohamed A, Amer E, Eldin N, Hossam M, Elmasry N, Adnan GT (2022) The impact of data processing and ensemble on breast cancer detection using deep learning. J Comput Commun 1(1):27–37. https://doi.org/10.21608/jocc.2022.218453
doi: 10.21608/jocc.2022.218453
Salvi M, Acharya UR, Molinari F, Meiburger KM (2021) The impact of pre- and post-image processing techniques on deep learning frameworks: a comprehensive review for digital pathology image analysis. Comput Biol Med 128:104129. https://doi.org/10.1016/j.compbiomed.2020.104129
doi: 10.1016/j.compbiomed.2020.104129 pubmed: 33254082
https://github.com/ehalle/Detecting_Ductopenic_Parotid_Salivary_Glands
Graham RNJ, Perriss RW, Scarsbrook AF (2005) DICOM demystified: a review of digital file formats and their use in radiological practice. Clin Radiol 60(11):1133–1140. https://doi.org/10.1016/j.crad.2005.07.003
doi: 10.1016/j.crad.2005.07.003 pubmed: 16223609
Frangi AF, Niessen WJ, Vincken KL, Viergever MA (1998) Multiscale vessel enhancement filtering. In: Proc 1st Medical Image Computing and Computer-Assisted Intervention. Springer, pp 130–137
Li X, Morgan PS, Ashburner J, Smith J, Rorden C (2016) The first step for neuroimaging data analysis: DICOM to NIfTI conversion. J Neurosci Methods 264:47–56. https://doi.org/10.1016/j.jneumeth.2016.03.001
doi: 10.1016/j.jneumeth.2016.03.001 pubmed: 26945974
Fishman EK, Ney DR, Heath DG, Corl FM, Horton KM, Johnson PT (2006) Volume rendering versus maximum intensity projection in CT angiography: what works best, when, and why. Radiographics 26(3):905–922. https://doi.org/10.1148/rg.263055186
doi: 10.1148/rg.263055186 pubmed: 16702462
Fontenele RC, Gerhardt MDN, Pinto JC, Van Gerven A, Willems H, Jacobs R, Freitas DQ (2022) Influence of dental fillings and tooth type on the performance of a novel artificial intelligence-driven tool for automatic tooth segmentation on CBCT images–a validation study. J Dent 119:104069. https://doi.org/10.1016/j.jdent.2022.104069
doi: 10.1016/j.jdent.2022.104069 pubmed: 35183696
Zhu X, Xue Z, Gao X, Zhu Y, Wong STC (2009) Voles: vascularity-oriented level set algorithm for pulmonary vessel segmentation in image guided intervention therapy. In: Proc of IEEE International Symposium on Biomedical Imaging: From Nano to Macro. pp 1247–1250 https://doi.org/10.1109/ISBI.2009.5193288
Jimenez-Carretero D, Santos A, Kerkstra S, Rudyanto RD, Ledesma-Carbayo MJ (2013) 3D Frangi-based lung vessel enhancement filter penalizing airways. In: Proc IEEE 10th International Symposium on Biomedical Imaging. pp 926–929 https://doi.org/10.1109/ISBI.2013.6556627
Lorenz C, Carlsen IC, Buzug TM, Fassnacht C, Weese J (1997) A multi-scale line filter with automatic scale selection based on the Hessian matrix for medical image segmentation. In: Proc Scale-Space Theory in Computer Vision. Springer, pp 2–163 https://doi.org/10.1007/3-540-63167-4_47
Zenk J, Hosemann WG, Iro H (1998) Diameters of the main excretory ducts of the adult human submandibular and parotid gland: a histologic study. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 85(5):576–580. https://doi.org/10.1016/S1079-2104(98)90294-3
doi: 10.1016/S1079-2104(98)90294-3 pubmed: 9619677
Mohammed R, Rawashdeh J, Abdullah M (2020) Machine learning with oversampling and undersampling techniques: overview study and experimental results. In: Proc IEEE 11th International Conference on Information and Communication Systems. pp 243–248 https://doi.org/10.1109/ICICS49469.2020.239556
Shorten C, Khoshgoftaar TM (2019) A survey on image data augmentation for deep learning. J Big Data 6(1):1–48. https://doi.org/10.1186/s40537-019-0197-0
doi: 10.1186/s40537-019-0197-0
Lin TY, Maire M, Belongie S, Hays J, Perona P, Ramanan D, Dollár P, Zitnick CL (2015) Microsoft COCO: common objects in context. In: Proc European Conference on Computer Vision. Springer, pp 740–755 https://doi.org/10.1007/978-3-319-10602-1_48
Walrand J (2021) Probability in electrical engineering and computer science: an application-driven course. Springer Nature. https://doi.org/10.1007/978-3-030-49995-2
Kise Y, Ikeda H, Fujii T, Fukuda M, Ariji Y, Fujita H et al (2019) Preliminary study on the application of deep learning system to diagnosis of Sjögren’s syndrome on CT images. Dentomaxillofac Radiol 48(6):20190019. https://doi.org/10.1259/dmfr.20190019
doi: 10.1259/dmfr.20190019 pubmed: 31075042 pmcid: 6747436
Kise Y, Shimizu M, Ikeda H, Fujii T, Kuwada C, Nishiyama M et al (2020) Usefulness of a deep learning system for diagnosing sjögren’s syndrome using ultrasonography images. Dentomaxillofac Radiol 49(3):20190348. https://doi.org/10.1259/dmfr.20190348
doi: 10.1259/dmfr.20190348 pubmed: 31804146 pmcid: 7068075

Auteurs

Elia Halle (E)

Department of Data Mining, Jerusalem College of Technology, Jerusalem, Israel.

Tevel Amiel (T)

Oral Maxillofacial Imaging Unit, Department of Oral Medicine, Sedation and Imaging, Faculty of Dental Medicine, The Hebrew University of Jerusalem, Hadassah Medical Center, Jerusalem, Israel.

Doron J Aframian (DJ)

Department of Oral Medicine, Sedation and Imaging, Faculty of Dental Medicine, The Hebrew University of Jerusalem, Hadassah Medical Center, Jerusalem, Israel.

Tal Malik (T)

Department of Data Mining, Jerusalem College of Technology, Jerusalem, Israel.

Avital Rozenthal (A)

Department of Data Mining, Jerusalem College of Technology, Jerusalem, Israel.

Oren Shauly (O)

School of Computer Science and Engineering, The Hebrew University of Jerusalem, Jerusalem, Israel.

Leo Joskowicz (L)

School of Computer Science and Engineering, The Hebrew University of Jerusalem, Jerusalem, Israel.

Chen Nadler (C)

Oral Maxillofacial Imaging Unit, Department of Oral Medicine, Sedation and Imaging, Faculty of Dental Medicine, The Hebrew University of Jerusalem, Hadassah Medical Center, Jerusalem, Israel.

Talia Yeshua (T)

Department of Data Mining, Jerusalem College of Technology, Jerusalem, Israel. yeshua@g.jct.ac.il.
Department of Applied Physics, Jerusalem College of Technology, Jerusalem, Israel. yeshua@g.jct.ac.il.

Classifications MeSH