Clinical decision support system in digital diaphanoscopy of maxillary sinuses based on convolutional neural networks

Bryanskaya E.O., Gerasin D.V., Bakotina A.V., Ovchinnikov A.Y., Nikolaeva Y.O., Maslennikova A.V., Dryomin, V.V., Dunaev, A.V.

For Russian citation (Opticheskii Zhurnal):

Брянская Е.О., Герасин Д.В., Бакотина А.В., Овчинников А.Ю., Николаева Ю.О., Масленникова А.В., Дрёмин В.В., Дунаев А.В. Система поддержки принятия врачебных решений в цифровой диафаноскопии верхнечелюстных пазух с применением сверточных нейронных сетей // Оптический журнал. 2026. Т. 93. № 5. С. 100–110. http://doi.org/ 10.17586/1023-5086-2026-93-05-100-110

Bryanskaya E.O., Gerasin D.V., Bakotina A.V., Ovchinnikov A.Yu., Nikolaeva Yu.O., Maslennikova A.V., Dremin V.V., Dunaev A.V. Clinical decision support system in digital diaphanoscopy of maxillary sinuses using convolutional neural networks [in Russian] // Opticheskii Zhurnal. 2026. V. 93. № 5. P. 100–110. http://doi.org/10.17586/1023-5086-2026-93-05-100-110

For citation (Journal of Optical Technology):

Abstract:

Subject of study. Methods and algorithms for supporting сlinical decision-making in digital diaphanoscopy of maxillary sinus using convolutional neural networks. Aim of study. Improving the accuracy of classification of maxillary sinus pathologies in digital diaphanoscopy using a classification model based on the convolutional neural network ResNet-50. Method. The registration of diaphanograms was carried out using the software-hardware system of digital diaphanoscopy at two probing wavelengths (650 and 850 nm). The analysis of the diaphanograms was carried out using the developed diaphanogram classification models based on convolutional neural network ResNet-50. Main results. The problem of differentiating maxillary sinus conditions into the classes “sinusitis”, “cystic changes” and “absence of pathology” has been solved using the developed clinical decision support system. The following accuracy indicators of the model were obtained: sensitivity 0.85, specificity 0.91, accuracy 0.84, balanced accuracy 0.88. To improve the effectiveness of classification, an upgraded model has been proposed, which makes it possible to achieve higher performance by applying different approaches to data generation, and consequently by increasing the number of them. Practical significance. The results allow to assess the prospects of using the developed classification model for the early detection of maxillary sinus pathologies in telemedicine and during automated consultations of otorhinolaryngologist using a clinical decision support system.

Keywords:

digital diaphanoscopy, clinical decision support system, maxillary sinuses, diaphanograms, convolutional neural networks

Acknowledgements:

the study was funded by Russian Science Foundation according to the research project № 24-75-00144

OCIS codes: 170.4580, 170.4940

References:

1. Вишняков В.В., Талалаев В.Н., Лукьянчиков В.А. и др. Пиоцеле agger nasi // Российская ринология. 2020. Т. 28. № 4. С. 241–245. https://doi.org/10.17116/rosrino202028041241

Vishnyakov V.V., Talalaev V.N., Lukyanchikov V.A., et al. Piocele agger nasi [in Russian] // Russian Rhinology. 2020. V. 28. № 4. P. 241–245. https://doi.org/10.17116/rosrino202028041241

2. Villarroel M., Blackwell D., Jen A. Tables of summary health statistics for U.S. Adults: 2018 National Health Interview Survey. National Center for Health Statistics, 2019. 9 p. [Electronic resource]. Access mode: https://archive.cdc.gov/#/details?url=https://www.cdc.gov/nchs/nhis/shs/tables.htm (accessed 02/09/2021).

3. Feldmann H. Die Geschichte der Diaphanoskopie [auf Deutsch] // Laryngo-Rhino-Otologie. 1998. V. 77. № 5. P. 297–304. https://doi.org/10.1055/s-2007-996978

4. Плужников М.C., Иванов Б.С., Усанов А.А. Лазерная диафаноскопия при воспалительных заболеваниях придаточных пазух носа // Вестник оториноларингологии. 1991. Т. 4. С. 22.

Pluzhnikov M.C., Ivanov B.S., Usanov A.A. Laser diaphanoscopy for inflammatory diseases of the paranasal sinuses [in Russian] // Vestnik otorinolaringologii. 1991. V. 4. P. 22.

5. Bryanskaya E.O., Dremin V.V., Shupletsov V.V., et al. Digital diaphanoscopy of maxillary sinus pathologies supported by machine learning // J. Biophotonics. 2023. V. 16. № 9. P. e202300138. https://doi.org/10.1002/jbio.20230013

6. Брянская Е.О., Дунаев А.В. Метод и устройство цифровой диафаноскопии для диагностики патологий верхнечелюстных пазух // Медицинская техника. 2023. № 3. С. 5–7.

Bryanskaya E.O., Dunaev A.V. A method and device for digital diaphanoscopy for the diagnosis of MS pathologies // Biomed Eng. 2023. V. 57. № 3. P. 159–163. https://doi.org/10.1007/s10527-023-10289-5

7. Gerasin D.V., Bryanskaya E.O., Dremin V.V., et al. The use of convolutional neural networks to classify the states of the maxillary sinuses in digital diaphanoscopy // 2024 Intelligent Technologies and Electronic Devices in Vehicle and Road Transport Complex (TIRVED). Moscow, Russia. November 13–15, 2024. P. 1–4. https://doi.org/10.1109/TIRVED63561.2024.10769806

8. Peters V.G., Wyman D.R., Patterson M.S., et al. Optical properties of normal and diseased human breast tissues in the visible and near infrared // Phys. Med. Biol. 1990. V. 35. № 9. P. 1317–1334. https://doi.org/10.1088/0031-9155/35/9/010

9. Van Veen R.L., Sterenborg H.J.C.M., Marinelli A.W.K.S., et al. Intraoperatively assessed optical properties of malignant and healthy breast tissue used to determine the optimum wavelength of contrast for optical mammography // J. Biomed. Opt. 2004. V. 9. № 6. P. 1129–1136. https://doi.org/10.1117/1.1803547

10. Jacques S.L. Optical properties of biological tissues: A review // Phys. Med. Biol. 2013. V. 58. № 11. P. R37–R61. https://doi.org/10.1088/0031-9155/58/11/R37

11. Bryanskaya E.O., Novikova I.N., Dremin V.V., et al. Optical diagnostics of the maxillary sinuses by digital diaphanoscopy technology // Diagnostics. 2021. V. 77. № 11. P.1–13. https://doi.org/10.3390/diagnostics11010077

12. Conigliaro J., Kapoor S. Screening/Physical exam/health maintenance / Handbook of Outpatient Medicine. Cham: Springer, 2023. P. 3–44. https://doi.org/10.1007/978-3-031-15353-2_1

13. Wu Q., Wang X., Liang G., et al. Advances in image-based artificial intelligence in otorhinolaryngology-head and neck surgery: A systematic review // Otolaryngol. — Head Neck Surg. 2023. V. 169. № 5. P. 1132–1142. https://doi.org/10.1002/ohn.391

14. Луцив В.Р. Сверточные искусственные нейронные сети глубокого обучения // Оптический журнал. 2015. Т. 82. № 8. С. 11–23.

Lutsiv V.R. Convolutional deep-learning artificial neural networks // J. Opt. Technol. 2015. V. 82. № 8. P. 499–508. https://doi.org/10.1364/JOT.82.000499

15. Bryanskaya E.O., Gneushev R.Yu., Novikova I.N., et al. Brightness controller optimization for the digital diaphanoscopy system // Proc. SPIE. 2021. V. 119191. P. 119191Z. https://doi.org/10.1117/12.2615012

16. Вакс В.Л., Домрачева Е.Г., Черняева М.Б. и др. Применение метода терагерцовой газовой спектроскопии высокого разрешения для анализа состава продуктов термического разложения тканей кист околоносовых пазух // Оптический журнал. 2021. Т. 88. № 3. С. 72–76. http://doi.org/10.17586/1023-5086-2021-88-03-72-76

Vaks V.L., Domracheva E.G., Chernyaeva M.B., et al. Application of high-resolution terahertz gas spectroscopy to the compositional analysis of the thermal decomposition products of paranasal sinus cyst tissue // J. Opt. Technol. 2021. V. 88. № 3. P. 166–168. https://doi.org/10.1364/JOT.88.000166

17. Chainansamit S., Chit-uea-ophat C., Reechaipichitkul W., et al. The diagnostic value of traditional nasal examination tools in an endoscopic era // Ear, Nose Throat J. 2021. V. 100. № 3. P. 167–171. https://doi.org/10.1177/0145561319875711

18. Valkov A., Nokolov G., Duhlenski B., et al. The application of ultrasound examination in the treatment of Acute Sinusitis. Comparing X-ray to ultrasound of paranasal sinuses // Int. Bull. Otorhinolaryngol. 2021. V. 17. № 1. P. 40–41. https://doi.org/10.14748/orl.v17i1.7750