The method for evaluating the symmetry of the globule pattern in artificial intelligence systems for the diagnosis of skin neoplasms

Methods for early non-invasive diagnosis of melanoma using computer vision systems are considered. Existing computer vision systems using neural networks for classifying dermoscopic images do not allow tracking which diagnostic features are used to assign images to a particular class, reducing physicians' trust in the results. As an alternative, an image analysis algorithm is proposed with the ability to present justifications for decisions made at each processing stage. The implementation of this algorithm is based on the medical algorithm of modified globular pattern analysis. A significant sign of malignancy in a neoplasm is its asymmetry. This criterion is widely used by doctors in visual assessment of skin neoplasms. However, currently, the issues of evaluating the symmetry of globular patterns in artificial intelligence systems are not fully studied and described. A method for evaluating the symmetry of globular patterns in artificial intelligence systems for diagnosing skin neoplasms has been developed. A dataset of dermoscopic images was formed, containing 50 images each of neoplasms with symmetrically and asymmetrically arranged globular patterns. Methods for isolating the neoplasm area and globules are described. A classification system based on a set of 12 quantitative symmetry characteristics has been developed. The Random Forest algorithm was used to classify images based on symmetry features. In the conducted experiment, a classification accuracy of 85% was achieved. The presented results contribute to the development of computer vision methods in dermatology and demonstrate the possibility of using the proposed method in clinical decision support systems for modified analysis of dermoscopic patterns for diagnosing skin neoplasms.

1. Kaprin A. D., Starinsky V. V., Shakhzadova A. O. (eds.): Malignant neoplasms in Russia in 2021 (morbidity and mortality). P. A. Hertsen Moscow Oncology Research Institute – branch of FSBI NMRRC of the Ministry of Health of Russia, Moscow (2022). (In Russ.)

2. Seeja R. D., Suresh A. Deep learning based skin lesion segmentation and classification of melanoma using support vector machine (SVM). Asian Pacific Journal of Cancer Prevention, 20(5), 1555–1561 (2019). https://doi.org/10.31557/APJCP.2019.20.5.1555

3. Nikitaev V. G., Pronichev A. N., Tamrazova O. B., Sergeev V. Yu., Selchuk V. Yu., Kozlov V. S., Lim A. O. Model for Estimating the Heterogeneity of the Distribution of Globule Characteristics in Images of Skin Neoplasms. Measurement Techniques, 64(9), 778–783 (2021). https://doi.org/10.1007/s11018-022-02003-w

4. Khismatullina Z. R., Chebotaryov V. V., Babenko E. A. Dermatoscopy in dermato-oncology: current state and perspectives. Creative Surgery and Oncology, 10(3), 241–248 (2020). (In Russ.) https://doi.org/10.24060/2076-3093-2020-10-3-241-248

5. Poudel B. R., Bhattarai B., Niraula P., Khanal A. A deep neural network solution for malignant melanoma detection. International Journal of Advanced Research and Publications, 3(11), 18–25 (2019). https://www.researchgate.net/publication/337649359

6. Sokolov D. V., Makhson A. N., Demidov L. V., Vorozhtsov G. N., Kuzmin S. G., Sokolov V. V. Dermatoscopy (epiluminescent superfcial microscopy): in vivo diagnosis of skin melanoma (literature review). Siberian Journal of Oncology, (5), 63–67 (2008). (In Russ.) https://elibrary.ru/jaqnbd

7. Dubensky V. V., Dubensky Vl. V. Skin neoplasms in the practice of dermatovenerologist. Issues of epidemiology, etiology and pathogenesis, diagnosis. Vestnik Dermatologii i Venerologii, (6), 22–40 (2008). (In Russ.) https://elibrary.ru/jvvocr

8. Kittler H., Rosendahl C., Cameron A., Tschandl P. Dermatoscopy: An algorithmic method based on pattern analysis. Maudrich, Vienna (2016).

9. Nikitaev V. G., Pronichev A. N., Nagornov O. V., Kruglova L. S., Sergeev V. Y., Otchenashenko A. I. An artificial intelligence model for the semantic segmentation of neoplasms on images of the skin. Biomedical Engineering, 58(1), 36–39 (2024). https://doi.org/10.1007/s10527-024-10361-8

10. Kittler H., Marghoob A. A., Argenziano G. et al. Standardization of terminology in dermoscopy/dermatoscopy: results of the third consensus conference of the international society of dermoscopy. Journal of the American Academy of Dermatology, 74(6), 1093–1106 (2016). https://doi.org/10.1016/j.jaad.2015.12.038

11. Lazarev Yu. D. Modern methods of diagnosis and treatment of melanoma. Fundamental Aspects of Mental Health, (3), 6–10 (2018). (In Russ.) https://elibrary.ru/ynrhnj

12. Celebi M. E., Kingravi H. A., Uddin B., Iyatomi, H., Aslandogan Y. A., Stoecker W. V., Moss R. H. A methodological approach to the classification of dermoscopy images. Computerized Medical Imaging and Graphics, 31(6), 362–373 (2007). https://doi.org/10.1016/j.compmedimag.2007.01.003

13. Argenziano G., Soyer H. P., Chimenti S. et al. Dermoscopy of pigmented skin lesions: results of a consensus meeting via the Internet. Journal of the American Academy of Dermatology, 48(5), 679–693 (2003). https://doi.org/10.1067/mjd.2003.281

14. Garnavi R., Aldeen M., Bailey J. Computer-aided diagnosis of melanoma using border- and wavelet-based texture analysis. IEEE Transactions on Information Technology in Biomedicine, 16(6), 1239–1252 (2012). https://doi.org/10.1109/TITB.2012.2212282

15. Betta G., Leo G. Di, Fabbrocini G., Paolillo A., Scalvenzi M. Automated Application of the “7-point checklist” Diagnosis Method for Skin Lesions: Estimation of Chromatic and Shape Parameters. 2005 IEEE Instrumentation and Measurement Technology Conference Proceedings, Ottawa, ON, Canada, 2005, pp. 1818–1822. https://doi.org/10.1109/IMTC.2005.1604486