A.N. Kalinichenko1, K.I. Agapova2, M.A. Bolozya3, K.K. Kukanov4, E.V. Sadykova5

1–3,5 St. Petersburg State Electrotechnical University «LETI» (St. Petersburg, Russia)
4 Polenov Neurosurgical Research Institute, branch of the Almazov National Medical Research Centre (St. Petersburg, Russia)
1 ank-bs@yandex.ru, 2 a-ksyusha@bk.ru, 3 sweet.cherry2552@gmail.com, 4 kukanov_kk@almazovcentre.ru, 5 elensadykova@yandex.ru

At the moment, despite extensive medical research on the causes of recurrence of meningioma of the brain, there are no generally accepted diagnostic standards and effective prognostic criteria that allow diagnosing these important manifestations at an early stage of development, which affects both the quality of life of patients who have undergone neurosurgical intervention and the mortality rate. One of the possible solutions is the use of machine learning methods that allow to build a software and algorithmic model capable of giving a probabilistic assessment of the development of relapse based on the analysis of existing indicators.

Objective – development and software implementation of a model for predicting the risk of meningioma recurrence, which makes it possible to create a software tool that helps neurosurgeons assess the risk of recurrence of brain malignancies in patients who have undergone neurosurgery to remove brain meningioma.

Based on the analysis of the set of diagnostic features presented by the doctors, a random forest model was chosen as the basic machine learning model. As a result of statistical analysis of the initial 160 features, the most informative of them (61 features) were selected and used as input data for the machine learning model. Using the Python programming language, a classifier model based on the random forest algorithm was implemented and optimized. The developed algorithm demonstrated a forecast accuracy of about 90% on the test data set.

Based on the proposed algorithm, a software package in the Python programming language has been developed, which can be used in practice as an additional source of information when making medical decisions by neurosurgeons.

Kalinichenko A.N., Agapova K.I., Bolozya M.A., Kukanov K.K., Sadykova E.V. Prediction of recurrences of cerebral meningioma based on the analysis of clinical signs using a random forest model. Biomedicine Radioengineering. 2025. V. 28. № 2. P. 14–19. DOI: https:// doi.org/10.18127/j15604136-202502-02 (In Russian)

1. Goldbrunner R., Stavrinou P., Jenkinson M.D., Sahm F. et al. EANO guideline on the diagnosis and management of meningiomas. Neuro oncol. 2021. № 23(11). P. 1821–1834. https://doi.org/10.1093/neuonc/noab150.
2. Kukanov K.K., Vorob'yova O.M., Zabrodskaya Yu.M., Potyomkina E.G., Ushanov V.V., Tastanbekov M.M., Ivanova N.E. Intrakranial'nye meningiomy: kliniko-intraskopicheskie i patomorfologicheskie prichiny recidivirovaniya s uchetom sovremennyh metodov lecheniya (obzor literatury). Sibirskij onkologicheskij zhurnal. 2022. № 21(4). S. 110–123 (In Russian).
3. Huntoon K., Toland A.M.S., Dahiya S. Meningioma: a review of clinicopathological and molecular aspects. Front Oncol. 2020.
   № 10(10). P. 1–14. https://doi.org/10.3389/fonc.2020.579599.
4. Danilov G.V., Shifrin M.A., Kotik K.V., Ishankulov T.A., Orlov Y.N., Kulikov A.S., Potapov A.A. Artificial intelligence in neurosurgery:
   A systematic review using topic modeling. Part I: Major research areas. Sovremennye Tehnologii v Medicine. 2020. V. 12(5). P. 1–112. https://doi.org/10.17691/stm2020.12.5.12 2.
5. Lemeshko B.Yu. Kriterii proverki otkloneniya raspredeleniya ot normal'nogo zakona (rukovodstvo po primeneniyu). Novosibirskij gosudarstvennyj tekhnicheskij universitet. 2014. S. 192.
6. Kras'ko O.V. Statisticheskij analiz dannyh v medicinskih issledovaniyah. Minsk: MGEU im. A.D. Saharova. 2014. S. 127 (In Russian).
7. Gadzhiev YA., SHalbuzova K.I. Primenenie metodov mashinnogo obucheniya v prognozirovanii i rannem obnaruzhenii raka. Sciences of Europe. 2022. № 108. S. 48 (In Russian).
8. Ob"yasnenie metoda sluchajnogo lesa [Elektronnyj resurs]. URL: https://nerdit.ru/obiasnieniie-mietoda-sluchainogho-liesa/ (In Russian).