V. G. Nikitayev1, A.N. Pronichev2, O.V. Nagornov3, V.Yu. Sergeev4, A.I. Otchenashenko5, A.A. Buslaev6

1–3, 5, 6 National Research Nuclear University “MEPhI” (Moscow, Russia)
2 Oncodermatology Clinic LLC (Moscow, Russia)

Melanoma is one of the deadliest forms of skin malignant tumors and its incidence has grown recently. One of the best countermeasures against skin oncological diseases is early diagnosis. Several diagnostic solutions based on artificial intelligence have been developed in order to increase quality and availability of skin tumors diagnostics, yet these solutions lack credibility from doctors as a result of their non-transparent algorithms. Recognition system of types of skin lesions based on renowned dermatoscopic method of pattern analysis (also known as “Chaos and Clues” algorithm) has a chance to increase doctors’ credibility to digital diagnostic instruments. Method of pattern analysis consists of several classification stages. The article represents a digital model of one stage of dermatoscopic method of pattern analysis, when the pattern of neoplasm consists only of lines. The aim of the work was to develop a recognition model of nucleoplasm’s type of lines pattern with the application of artificial intelligence. The pattern may represent reticular, spread, parallel or curved lines. As a result, a model based on pretrained convolutional neural network has been developed with quality metrics: accuracy = 72%, average sensitivity = 0,72, average specificity = 0,91. The model is designed to be used as a part of recognition system of types of skin lesions based on dermatoscopic method of pattern analysis.

Nikitayev V.G., Pronichev A.N., Nagornov O.V., Sergeev V.Yu., Otchenashenko A.I., Buslaev A.A. Digital oncodermatology: recognition model of nucleoplasm’s type of lines pattern in lack of other features based on neural networks. Biomedicine Radioengineering. 2025. V. 28. № 2. P. 26–32. DOI: https:// doi.org/10.18127/j15604136-202502-04 (In Russian)

1. Melanoma v cifrah. GBUZ «Buryatskij respublikanskij klinicheskij onkologicheskij dispanser» URL: https://brkod.ru/news/melanoma-v-tsifrakh (data obrashcheniya: 11.01.2024) (In Russian).
2. Erkenova F.D., Puzin S.N. Statistika melanomy v Rossii i stranah Evropy. Mediko-social'naya ekspertiza i reabilitaciya. 2020. № 1. S. 44–52 (In Russian).
3. Rigel D.S., Garucci J.A. Malignant Melanoma: Prevention, Early Detection, and Treatment in the 21st Century. CA Cancer J Clin. 2000 V. 50. № 4. P. 215–236.
4. Tschandl P., Codella N., Akay B.N., Argenziano G., Braun R.P., Cabo H., Gutman D., Halpern A., Helba B., Hofmann-Wellenhof R., Lallas A., Lapins J., Longo C., Malvehy J., Marchetti M.A., Marghoob A., Menzies S., Oakley A., Paoli J., Puig S., Rinner C., Rosendahl C., Scope A., Sinz C., Soyer H.P., Thomas L., Zalaudek I., Kittler H. Comparison of the accuracy of human readers versus machine-learning algorithms for pigmented skin lesion classification: an open, web-based, international, diagnostic study. Lancet Oncol. 2019. V. 20. № 7. P. 938–947.
5. Soenksen L.R., Kassis T., Conover S.T., Marti-Fuster B., Birkenfeld J.S., Tucker-Schwartz J., Naseem A., Stavert R.R., Kim C.C., Senna M.M., Avilés-Izquierdo J., Collins J.J., Barzilay R., Gray M.L. Using deep learning for dermatologist-level detection of suspicious pigmented skin lesions from wide-field images. Sci Transl Med. 2021. V. 13. № 581. P. 1–12.
6. Haenssle H.A., Fink C., Schneiderbauer R., Toberer F., Buhl T., Blum A., Kalloo A., Hassen A.B.H., Thomas L., Enk A., Uhlmann L.; Reader study level-I and level-II Groups; Alt C., Arenbergerova M., Bakos R., Baltzer A., Bertlich I., Blum A., Bokor-Billmann T., Bowling J., Braghiroli N., Braun R., Buder-Bakhaya K., Buhl T., Cabo H., Cabrijan L., Cevic N., Classen A., Deltgen D., Fink C., Georgieva I., Hakim-Meibodi L.E., Hanner S., Hartmann F., Hartmann J., Haus G., Hoxha E., Karls R., Koga H., Kreusch J., Lallas A., Majenka P., Marghoob A., Massone C., Mekokishvili L., Mestel D., Meyer V., Neuberger A., Nielsen K., Oliviero M., Pampena R., Paoli J., Pawlik E., Rao B., Rendon A., Russo T., Sadek A., Samhaber K., Schneiderbauer R., Schweizer A., Toberer F., Trennheuser L., Vlahova L., Wald A., Winkler J., Wölbing P., Zalaudek I. Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists. Ann Oncol. 2018. V. 29. № 8. P. 1836–1842.
7. Zhuchkov M. V., Bulin'ska A. K., Kittler G. Primenenie algoritma «Haos i Priznaki» v ocenke dermatoskopicheskih izobrazhenij pigmentnyh novoobrazovanij kozhi. Dermatologiya. Prilozhenie k zhurnalu Consilium medicum. 2017. № 2. S. 5–13 (In Russian).
8. Nikitaev V.G., Pronichev A.N., Nagornov O.V., Kruglova L.S., Sergeev V.Yu., Otchenashenko A.I. Model' iskusstvennogo intellekta dlya semanticheskoj segmentacii novoobrazovanij na izobrazheniyah kozhi. Medicinskaya tekhnika. 2024. № 1 (343). S. 26–28 (In Russian).
9. Nikitaev V.G., Pronichev A.N., Nagornov O.V., Sergeev V.Yu., Kruglova L.S., Otchenashenko A.I., Deeva O.K. Metod ocenki simmetrii uzora globul v sistemah iskusstvennogo intellekta dlya diagnostiki novoobrazovanij kozhi. Izmeritel'naya tekhnika. 2024. № 73(9). S. 53–60 (In Russian).
10. Okunev S.V., Favorskaya M.N. Primenenie augmentacii i generativno-sostyazatel'noj nejronnoj seti dlya uvelicheniya naborov dannyh. Aktual'nye problemy aviacii i kosmonavtiki. 2020. T. 2. S. 162–164 (In Russian).
11. Shorten C., Khoshgoftaar T.M. A survey on Image Data Augmentation for Deep Learning. J. Big Data. 2019. V. 6. P. 1–48.
12. Kozachok A.V., Spirin A.A., Kozachok E.S. Obzor metodov rannego obnaruzheniya melanomy s ispol'zovaniem metodov komp'yuternogo zreniya. Trudy ISP RAN. 2022. T. 34. № 4. S. 241–250 (In Russian).
13. Shchetinin E.Yu., Sevast'yanov L.A. O metodah perenosa glubokogo obucheniya v zadachah klassifikacii biomedicinskih izobrazhenij. Informatika i ee primeneniya. 2021. T. 15. № 4. S. 59–64 (In Russian).
14. Nejronnye seti predpochitayut tekstury i kak s etim borot'sya. Habr URL: https://habr.com/ru/companies/ods/articles/453788/ (data obrashcheniya: 11.03.2023) (In Russian).
15. How to explain the ROC curve and ROC AUC score?. Evidently AI URL: https://www.evidentlyai.com/classification-metrics/explain-roc-curve (дата обращения: 11.03.2023).