Segmentation of 3D OCT Images of Human Skin Using Neural Networks with U-Net Architecture

The aim of the study is a comparative analysis of algorithms for segmentation of three-dimensional OCT images of human skin using neural networks based on U-Net architecture when training the model on two-dimensional and three-dimensional data.

Materials and Methods. Two U-Net-based network architectures for segmentation of 3D OCT skin images are proposed in this work, in which 2D and 3D blocks of 3D images serve as input data. Training was performed on thick skin OCT images acquired from 7 healthy volunteers. For training, the OCT images were semi-automatically segmented by experts in OCT and dermatology. The Sørensen–Dice coefficient, which was calculated from the segmentation results of images that did not participate in the training of the networks, was used to assess the quality of segmentation. Additional testing of the networks’ capabilities in determining skin layer thicknesses was performed on an independent dataset from 8 healthy volunteers.

Results. In evaluating the segmentation quality, the values of the Sørensen–Dice coefficient for the upper stratum corneum, ordered stratum corneum, epidermal cellular layer, and dermis were 0.90, 0.94, 0.89, and 0.99, respectively, for training on two-dimensional data and 0.89, 0.94, 0.87, and 0.98 for training on three-dimensional data. The values obtained for the dermis are in good agreement with the results of other works using networks based on the U-Net architecture. The thicknesses of the ordered stratum corneum and epidermal cellular layer were 153±24 and 137±17 μm, respectively, when the network was trained on two-dimensional data and 163±19 and 137±20 μm when trained on three-dimensional data.

Conclusion. Neural networks based on U-Net architecture allow segmentation of skin layers on OCT images with high accuracy, which makes these networks promising for obtaining valuable diagnostic information in dermatology and cosmetology, e.g., for estimating the thickness of skin layers.

1. Rukovodstvo po opticheskoy kogerentnoy tomografii [Handbook of optical coherence tomography]. Pod red. Gladkovoy N.D., Shakhovoy N.M., Sergeeva A.M. [Gladkova N.D., Shakhova N.M., Sergeev A.M. (editors)]. Moscow: Fizmatlit; 2007.
2. Mul’timodal’naya opticheskaya kogerentnaya tomografiya v klinicheskoy meditsine [Multimodal optical coherence tomography in clinical medicine]. Pod red. Gladkovoy N.D., Gelikonova G.V., Kiselevoy E.B. [Gladkova N.D., Gelikonov G.V., Kiseleva E.B. (editors)]. Moscow: Fizmatlit; 2022.
3. Wan B., Ganier C., Du-Harpur X., Harun N., Watt F.M., Patalay R., Lynch M.D. Applications and future directions for optical coherence tomography in dermatology. Br J Dermatol 2021; 184(6): 1014–1022, https://doi.org/10.1111/bjd.19553.
4. Venhuizen F.G., van Ginneken B., Liefers B., van Grinsven M.J.J.P., Fauser S., Hoyng C., Theelen T., Sánchez C.I. Robust total retina thickness segmentation in optical coherence tomography images using convolutional neural networks. Biomed Opt Express 2017; 8(7): 3292–3316, https://doi.org/10.1364/BOE.8.003292.
5. Zhang X., Yousefi S., An L., Wang R.K. Automated segmentation of intramacular layers in Fourier domain optical coherence tomography structural images from normal subjects. J Biomed Opt 2012; 17(4): 046011, https://doi.org/10.1117/1.JBO.17.4.046011.
6. Schmitt J.M., Xiang S.H., Yung K.M. Speckle in optical coherence tomography. J Biomed Opt 1999; 4(1): 95–105, https://doi.org/10.1117/1.429925.
7. Hori Y., Yasuno Y., Sakai S., Matsumoto M., Sugawara T., Madjarova V., Yamanari M., Makita S., Yasui T., Araki T., Itoh M., Yatagai T. Automatic characterization and segmentation of human skin using three-dimensional optical coherence tomography. Opt Express 2006; 14(5): 1862–1877, https://doi.org/10.1364/oe.14.001862.
8. Ali M., Hadj B. Segmentation of oct skin images by classification of speckle statistical parameters. IEEE International Conference on Image Processing 2010; p. 613–616, https://doi.org/10.1109/icip.2010.5653019.
9. Li A., Cheng J., Yow A.P., Wall C., Wong D.W., Tey H.L., Liu J. Epidermal segmentation in high-definition optical coherence tomography. Annu Int Conf IEEE Eng Med Biol Soc 2015; 2015: 3045–3048, https://doi.org/10.1109/EMBC.2015.7319034.
10. Taghavikhalilbad A., Adabi S., Clayton A., Soltanizadeh H., Mehregan D., Avanaki M.R.N. Semi-automated localization of dermal epidermal junction in optical coherence tomography images of skin. Appl Opt 2017; 56(11): 3116–3121, https://doi.org/10.1364/AO.56.003116.
11. Srivastava R., Yow A.P., Cheng J., Wong D.W.K., Tey H.L. Three-dimensional graph-based skin layer segmentation in optical coherence tomography images for roughness estimation. Biomed Opt Express 2018; 9(8): 3590–3606, https://doi.org/10.1364/BOE.9.003590.
12. Ronneberger O., Fischer P., Brox T. U-Net: convolutional networks for biomedical image segmentation. Medical image computing and computer-assisted intervention — MICCAI 2015. 2015; p. 234–241, https://doi.org/10.1007/978-3-319-24574-4_28.
13. Roy A.G., Conjeti S., Karri S.P.K., Sheet D., Katouzian A., Wachinger C., Navab N. ReLayNet: retinal layer and fluid segmentation of macular optical coherence tomography using fully convolutional networks. Biomed Opt Express 2017; 8(8): 3627–3642, https://doi.org/10.1364/BOE.8.003627.
14. Calderon-Delgado M., Tjiu J.W., Lin M.Y., Huang S.L. High resolution human skin image segmentation by means of fully convolutional neural networks. 2018 International Conference on Numerical Simulation of Optoelectronic Devices (NUSOD). 2018; p. 31–32, https://doi.org/10.1109/nusod.2018.8570241.
15. Kepp T., Droigk C., Casper M., Evers M., Hüttmann G., Salma N., Manstein D., Heinrich M.P., Handels H. Segmentation of mouse skin layers in optical coherence tomography image data using deep convolutional neural networks. Biomed Opt Express 2019; 10(7): 3484–3496, https://doi.org/10.1364/BOE.10.003484.
16. Chueh K.M., Kao H.L., Chen H.H., Shun C.T., Calderon-Delgado M., Huang S.L. Deep feature learning for contour segmentation of aorta’s intima by using sub-micron-resolution OCT. 2019 IEEE International Conference on BioPhotonics (BioPhotonics). 2019; p. 1–2, https://doi.org/10.1109/biophotonics.2019.8896753.
17. Gu Z., Cheng J., Fu H., Zhou K., Hao H., Zhao Y., Zhang T., Gao S., Liu J. CE-Net: context encoder network for 2D medical image segmentation. IEEE Trans Med Imaging 2019; 38(10): 2281–2292, https://doi.org/10.1109/TMI.2019.2903562.
18. Del Amor R., Morales S., Colomer A., Mogensen M., Jensen M., Israelsen N.M., Bang O., Naranjo V. Automatic segmentation of epidermis and hair follicles in optical coherence tomography images of normal skin by convolutional neural networks. Front Med (Lausanne) 2020; 7: 220, https://doi.org/10.3389/fmed.2020.00220.
19. Czajkowska J., Badura P., Korzekwa S., Płatkowska-Szczerek A. Deep learning approach to skin layers segmentation in inflammatory dermatoses. Ultrasonics 2021; 114: 106412, https://doi.org/10.1016/j.ultras.2021.106412.
20. Liu X., Chuchvara N., Liu Y., Rao B. Real-time deep learning assisted skin layer delineation in dermal optical coherence tomography. OSA Contin 2021; 4(7): 2008–2023, https://doi.org/10.1364/osac.426962.
21. Gao T., Liu S., Gao E., Wang A., Tang X., Fan Y. Automatic segmentation of laser-induced injury OCT images based on a deep neural network model. Int J Mol Sci 2022; 23(19): 11079, https://doi.org/10.3390/ijms231911079.
22. Ji Y., Yang S., Zhou K., Lu J., Wang R., Rocliffe H.R., Pellicoro A., Cash J.L., Li C., Huang Z. Semisupervised representative learning for measuring epidermal thickness in human subjects in optical coherence tomography by leveraging datasets from rodent models. J Biomed Opt 2022; 27(8): 085002, https://doi.org/10.1117/1.JBO.27.8.085002.
23. Zhang Y., Liao Q., Ding L., Zhang J. Bridging 2D and 3D segmentation networks for computation efficient volumetric medical image segmentation: an empirical study of 2.5D solutions. arXiv 2022, https://doi.org/10.48550/arXiv.2010.06163.
24. Shlivko I.L., Kirillin M.Y., Donchenko E.V., Ellinsky D.O., Garanina O.E., Neznakhina M.S., Agrba P.D., Kamensky V.A. Identification of layers in optical coherence tomography of skin: comparative analysis of experimental and Monte Carlo simulated images. Skin Res Technol 2015; 21(4): 419–425, https://doi.org/10.1111/srt.12209.
25. Kirillin M., Meglinski I., Kuzmin V., Sergeeva E., Myllylä R. Simulation of optical coherence tomography images by Monte Carlo modeling based on polarization vector approach. Opt Express 2010; 18(21): 21714–21724, https://doi.org/10.1364/OE.18.021714.
26. Kurakina D., Sergeeva E., Khilov A., Kirillin M. Light dose and fluorescence imaging depth in dual-wavelength PDT: a numerical study for various photosensitizer distributions in a layered biotissue. Journal of Biomedical Photonics & Engineering 2024; 10(4): 040318, https://doi.org/10.18287/jbpe24.10.040318.

Shishkova V.A., Gromov N.V., Mironycheva A.M., Kirillin M.Yu. Segmentation of 3D OCT Images of Human Skin Using Neural Networks with U-Net Architecture. Sovremennye tehnologii v medicine 2025; 17(1): 6, https://doi.org/10.17691/stm2025.17.1.01