CC BY
12*
ALT'23 The 30th International Conference on Advanced Laser Technologies
B-P-6
Multivariate analysis of Raman spectra and dematoscopic images for
the diagnosis of skin cancer
I. Matveeva1, V. Derugina1, L. Bratchenko1, Y. Khristoforova1, I. Bratchenko1, V. Zakharov1
1- Samara National Research University named after academician S.P. Korolev, 34, Moskovskoye shosse, Samara,
443086, Russia
m-irene-a@yandex. ru
The problems of diagnosing oncological diseases are associated with their wide distribution and a steady upward trend. Cancer research plays a vital role in the non-invasive diagnosis, staging and monitoring of various types of cancer and typically involves sophisticated instrumentation to provide detailed information about tumor size and location. Accurate diagnosis of the cancer type and early diagnosis are key factors in the successful treatment of neoplasms. Recently, for the diagnosis of skin cancer, it has been proposed to use dermatoscopy, which makes it possible to detect and visualize surface heterogeneities [2]. The modern trend in the development of optical diagnostics is also the use of Raman spectroscopy (RS) methods, which actually implement the tissue "optical biopsy" by identifying the chemical features of the tumor based on spectral data [1]. The application of machine learning methods makes it possible to ensure high accuracy of RS diagnostics of malignant neoplasms (more than 90%) [3], however, it has the same disadvantages of pinpoint biopsy (the possibility of an erroneous diagnosis if the optical biopsy site is chosen incorrectly) and reduced efficiency in multiclass diagnostics of the cancer type prevents real-time visualization of the tumor. The paper is devoted to developing combined method for diagnosing skin cancer based on Raman spectra and dermatoscopic images of skin tissue. An in vivo study of skin tumors was carried out at the Samara Regional Clinical Oncology Center. The study involved 540 patients. Experimental skin Raman spectra were recorded using a portable setup that includes a laser source with a central wavelength of 785 nm [4]. The spectra were recorded with a spectral resolution of 0.2 nm in the range from 837 to 920 nm, which corresponds to 792-1874 cm-1. Raman spectra of a healthy skin area and skin neoplasms were recorded in each patient. A total of 1000 spectra were used: 540 healthy skin, 113 keratosis, 122 basal cell carcinoma, 67 malignant melanoma, and 158 pigmented nevus spectra. Dermatoscopic images of this data set were obtained using a prototype multispectral digital dermatoscope [5]. The dataset consists of a total of 314 images: 104 malignant melanoma, 200 pigmented nevi.
Machine learning methods, in particular, convolutional neural networks, were used to analyze the registered data. Classification models for the main diagnostic cases have shown an increase in classification accuracy compared to the analysis of Raman spectra or dermatoscopic images alone. As a result, combined method for diagnosing skin cancer, which simultaneously takes into account both specific spectral features of tumors and spatial inhomogeneities in the distribution of optical density, has been proposed. The studied approaches to the analysis of spectral data can be further used as part of the software for automated screening diagnostics of skin pathologies in order to detect tumors at an early stage of development.
[1] J. Popp, C. Krafft and T. Mayerhöfer, Modern Raman spectroscopy for biomedical applications: A variety of Raman spectroscopical techniques on the threshold of biomedical applications, Optik & Photonik, vol. 6, №4, 24-28, (2011).
[2] V.A. Deryugina, I.A. Matveeva, I.A. Bratchenko, Neural network classification of dermatological images, International scientific forum on control and engineering, pp. 39-41, (2022).
[3] I.A. Bratchenko, L.A. Bratchenko, Y.A. Khristoforova, A.A. Moryatov, S.V. Kozlov and V.P. Zakharov, Classification of skin cancer using convolutional neural networks analysis of Raman spectra, Computer Methods and Programs in Biomedicine, vol. 219, pp. 106755, (2022).
[4] I.A. Bratchenko, L.A. Bratchenko, A.A. Moryatov, Y.A. Khristoforova, D.N. Artemyev, O.O. Myakinin, A.E. Orlov, S.V. Kozlov and V.P. Zakharov, In vivo diagnosis of skin cancer with a portable Raman spectroscopic device, Experimental Dermatology, vol. 30, N°5, pp. 652-663, (2021).
[5] S.G. Konovalov, O.A. Melsitov, O.O. Myakinin, I. A. Bratchenko, A.A. Moryatov, S.V. Kozlov and V. P. Zakharov, Dermatoscopy software tool for in vivo automatic malignant lesions detection, Journal of Biomedical Photonics & Engineering, vol. 4, N°4, pp. 040302, (2018).
