<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE root>
<article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:ali="http://www.niso.org/schemas/ali/1.0/" article-type="other" dtd-version="1.2" xml:lang="en"><front><journal-meta><journal-id journal-id-type="publisher-id">Digital Diagnostics</journal-id><journal-title-group><journal-title xml:lang="en">Digital Diagnostics</journal-title><trans-title-group xml:lang="ru"><trans-title>Digital Diagnostics</trans-title></trans-title-group><trans-title-group xml:lang="zh"><trans-title>Digital Diagnostics</trans-title></trans-title-group></journal-title-group><issn publication-format="print">2712-8490</issn><issn publication-format="electronic">2712-8962</issn><publisher><publisher-name xml:lang="en">Eco-Vector</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="publisher-id">688346</article-id><article-id pub-id-type="doi">10.17816/DD688346</article-id><article-categories><subj-group subj-group-type="toc-heading" xml:lang="en"><subject>Original Study Articles</subject></subj-group><subj-group subj-group-type="toc-heading" xml:lang="ru"><subject>Оригинальные исследования</subject></subj-group><subj-group subj-group-type="toc-heading" xml:lang="zh"><subject>原创性科研成果</subject></subj-group><subj-group subj-group-type="article-type"><subject></subject></subj-group></article-categories><title-group><article-title xml:lang="en">Application of radiomics in differential diagnostics of malignant and benign ovarian tumors</article-title><trans-title-group xml:lang="ru"><trans-title>Применение радиомики в дифференциальной диагностике злокачественных и доброкачественных образований яичников</trans-title></trans-title-group><trans-title-group xml:lang="zh"><trans-title/></trans-title-group></title-group><contrib-group><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0001-5994-0468</contrib-id><contrib-id contrib-id-type="spin">3018-2527</contrib-id><name-alternatives><name xml:lang="en"><surname>Nudnov</surname><given-names>Nikolay V.</given-names></name><name xml:lang="ru"><surname>Нуднов</surname><given-names>Николай Васильевич</given-names></name><name xml:lang="zh"><surname>Nudnov</surname><given-names>Nikolay V.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD, Dr. Sci. (Medicine), Professor</p></bio><bio xml:lang="ru"><p>д-р мед. наук, профессор</p></bio><bio xml:lang="zh"><p>MD, Dr. Sci. (Medicine), Professor</p></bio><email>nudnov@rncrr.ru</email><xref ref-type="aff" rid="aff1"/><xref ref-type="aff" rid="aff2"/><xref ref-type="aff" rid="aff3"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0003-2552-5754</contrib-id><contrib-id contrib-id-type="spin">4858-4627</contrib-id><name-alternatives><name xml:lang="en"><surname>Aksenova</surname><given-names>Svetlana P.</given-names></name><name xml:lang="ru"><surname>Аксенова</surname><given-names>Светлана Павловна</given-names></name><name xml:lang="zh"><surname>Aksenova</surname><given-names>Svetlana P.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>Ph.D. in Medicine, researcher</p></bio><bio xml:lang="ru"><p>канд. мед. наук, научный сотрудник</p></bio><bio xml:lang="zh"><p>Ph.D. in Medicine, researcher</p></bio><email>fabella@mail.ru</email><xref ref-type="aff" rid="aff4"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0004-2560-4879</contrib-id><contrib-id contrib-id-type="spin">2047-5006</contrib-id><name-alternatives><name xml:lang="en"><surname>Kuznetsova</surname><given-names>Darya D.</given-names></name><name xml:lang="ru"><surname>Кузнецова</surname><given-names>Дарья Дмитриевна</given-names></name><name xml:lang="zh"><surname>Kuznetsova</surname><given-names>Darya D.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><email>sssdasha@mail.ru</email><xref ref-type="aff" rid="aff5"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0007-9837-1983</contrib-id><name-alternatives><name xml:lang="en"><surname>Gribanov</surname><given-names>Nikita A.</given-names></name><name xml:lang="ru"><surname>Грибанов</surname><given-names>Никита Александрович</given-names></name><name xml:lang="zh"><surname>Gribanov</surname><given-names>Nikita A.</given-names></name></name-alternatives><email>Griboeshkanikita@gmail.com</email><xref ref-type="aff" rid="aff5"/></contrib></contrib-group><aff-alternatives id="aff1"><aff><institution xml:lang="en">Russian Scientific Center of Roentgenoradiology</institution></aff><aff><institution xml:lang="ru">Российский научный центр рентгенорадиологии</institution></aff><aff><institution xml:lang="zh">Russian Scientific Center of Roentgenoradiology</institution></aff></aff-alternatives><aff-alternatives id="aff2"><aff><institution xml:lang="en">Russian Medical Academy of Continuous Professional Education</institution></aff><aff><institution xml:lang="ru">Российская медицинская академия непрерывного профессионального образования</institution></aff><aff><institution xml:lang="zh">Russian Medical Academy of Continuous Professional Education</institution></aff></aff-alternatives><aff-alternatives id="aff3"><aff><institution xml:lang="en">Peoples’ Friendship University of Russia</institution></aff><aff><institution xml:lang="ru">Российский университет дружбы народов имени Патриса Лумумбы</institution></aff><aff><institution xml:lang="zh">Peoples’ Friendship University of Russia</institution></aff></aff-alternatives><aff-alternatives id="aff4"><aff><institution xml:lang="en">Russian Scientific Center of Roentgenoradiology, RUDN University</institution></aff><aff><institution xml:lang="ru">Российский научный центр рентгенорадиологии, Российский университет дружбы народов</institution></aff><aff><institution xml:lang="zh">Russian Scientific Center of Roentgenoradiology, RUDN University</institution></aff></aff-alternatives><aff-alternatives id="aff5"><aff><institution xml:lang="en">Russian Scientific Center of Roentgenoradiology</institution></aff><aff><institution xml:lang="ru">Федеральное государственное бюджетное учреждение "Российский научный центр рентгенорадиологии" Министерства здравоохранения Российской Федерации (ФГБУ "РНЦРР" Минздрава России)</institution></aff><aff><institution xml:lang="zh">Russian Scientific Center of Roentgenoradiology</institution></aff></aff-alternatives><pub-date date-type="preprint" iso-8601-date="2026-04-08" publication-format="electronic"><day>08</day><month>04</month><year>2026</year></pub-date><volume>7</volume><issue>1</issue><issue-title xml:lang="ru"/><history><date date-type="received" iso-8601-date="2025-07-29"><day>29</day><month>07</month><year>2025</year></date><date date-type="accepted" iso-8601-date="2026-03-12"><day>12</day><month>03</month><year>2026</year></date></history><permissions><copyright-statement xml:lang="en">Copyright ©; , Eco-Vector</copyright-statement><copyright-statement xml:lang="ru">Copyright ©; , Эко-вектор</copyright-statement><copyright-statement xml:lang="zh">Copyright ©; , Eco-Vector</copyright-statement><copyright-holder xml:lang="en">Eco-Vector</copyright-holder><copyright-holder xml:lang="ru">Эко-вектор</copyright-holder><copyright-holder xml:lang="zh">Eco-Vector</copyright-holder><ali:free_to_read xmlns:ali="http://www.niso.org/schemas/ali/1.0/"/><license><ali:license_ref xmlns:ali="http://www.niso.org/schemas/ali/1.0/">https://creativecommons.org/licenses/by-nc-nd/4.0</ali:license_ref></license></permissions><self-uri xlink:href="https://jdigitaldiagnostics.com/DD/article/view/688346">https://jdigitaldiagnostics.com/DD/article/view/688346</self-uri><abstract xml:lang="en"><p><bold>Background: </bold>Ovarian diseases are a serious problem for women's health. According to WHO, ovarian cancer is the 8th most common cancer and is one of the leading causes of death among oncological diseases in women.</p> <p> </p><p> </p><p>Traditional imaging methods are widely used but have limitations in tumor differential diagnosis. Existing stratification systems (e.g., O-RADS) do not always accurately assess malignancy risk, complicating treatment decisions.</p>   <p>Radiomics is a new technology that can improve the accuracy of ovarian tumor diagnosis compared to traditional imaging methods, making it a promising tool for the differential diagnosis of benign and malignant tumors.</p> <p><bold>Aim:</bold> To identify significant criteria (radiomic markers) based on radiomic analysis for predicting the malignant potential of ovarian lesions.<bold>Methods:</bold> A retrospective single-center randomized study was conducted with analysis of data from 304 patients. Taking into account the exclusion criteria, the final sample included 100 patients: 50 with histologically confirmed ovarian adenocarcinoma and 50 with histologically and/or clinically confirmed ovarian cystadenoma.</p> <p>Segmentation of regions of interest was performed by two independent researchers on T1 and T2 sequences in 3D-Slicer software. For each formation, 107 radiomic indicators were calculated from T1 and T2 sequences.</p> <p>The data were divided into training and testing sets (ratio 80%:20%). Feature reduction was performed using statistical and correlation analysis and feature significance assessment based on the Lasso model. Machine learning models were built in Python 3.12 programming language using specialized libraries. The selection of the model hyperparameters was performed using the GridSearchSV grid search algorithm using cross-validation. The effectiveness of the models was assessed using such metrics as the concordance index, accuracy and area under the ROC-AUC characteristic curve, and the confusion matrix.</p> <p><bold>Results: </bold>For the T1-based radiomic model, the LASSO model incorporating 16 predictive features demonstrated optimal performance: ROC-AUC 0.98, accuracy 0.98, C-index 1.624. For the T2-based model, the LASSO model with 18 features achieved: ROC-AUC 0.97, accuracy 0.91, C-index 0.616. Both models exhibited high sensitivity and specificity.</p> <p><bold>Conclusion:</bold> Radiomic markers extracted from T1- and T2-weighted sequences show significant potential for reliable differential diagnosis of benign and malignant ovarian tumors.</p></abstract><trans-abstract xml:lang="ru"><p><bold>Обоснование. </bold>Заболевания яичников представляют серьёзную проблему для женского здоровья. По данным ВОЗ, рак яичников занимает 8-е место по частоте и является одной из главных причин смертности среди онкологических заболеваний у женщин.</p> <p>Традиционные методы визуализации широко применяются, но имеют ряд ограничений в дифференциальной диагностике опухолей. А существующие стратификационные системы (например, O-RADS) не всегда позволяют точно определить риск злокачественности, что затрудняет выбор оптимальной лечебной тактики. </p> <p>Радиомика — новая технология, использование которой может потенциально повысить точность диагностики опухолей яичника по сравнению с традиционными методами визуализации, поэтому рассматривается как перспективный инструмент для дифференциальной диагностики доброкачественных и злокачественных образований.</p> <p><bold>Цель.</bold> Выявить значимые критерии (радиомические маркеры) на основе радиомического анализа, позволяющие спрогнозировать злокачественный потенциал серозных образований яичников. </p> <p><bold>Методы.</bold></p> <p>Проведено ретроспективное одноцентровое выборочное исследование с анализом данных 304 пациентов. С учетом критериев исключения в окончательную выборку вошли 106 пациентов: 53 с аденокарциномой яичника, подтвержденной гистологически, и 53 с цистаденомой яичника, подтвержденной гистологически и/или клинически (соотношение 1:1).</p> <p>Сегментация областей интереса выполнялась двумя независимыми исследователями на Т1- и Т2- последовательностях в программном обеспечении 3D-Slicer. Для каждого образования было рассчитано по 107 радиомических показателей из Т1- и Т2- последовательностей. </p> <p>Данные были разделены на обучающую и тестовую выборки ( соотношение 80%:20%). Проводились сокращение признаков с использованием статистического и корреляционного анализа и оценка значимости признаков на основе модели Lasso. Построение моделей машинного обучения проводилась на языке программирования Python 3.12 с использованием специализированных библиотек. Подбор гиперпараметров модели производился с помощью алгоритма поиска по сетке GridSearchSV с использованием кросс-валидации. Эффективность моделей оценивали по таким метрикам, как индекс конкордантности (С-индекс), accuracy и площадь под характеристической кривой ROC-AUC, матрица ошибок (confusion matrix).</p> <p><bold>Результаты. </bold></p> <p>Для модели, основанной на радиомических признаках, извлеченных с Т1-последовательности, наиболее эффективный результат показала модель LASSO, включающая 16 прогностических признаков. ROC-AUC: 0.98, accuracy: 0.98, C-индекс: 1.624. Для модели, основанной на радиомических признаках, извлеченных с Т2-последовательности, наиболее эффективный результат показала модель Lasso, включающая 18 прогностических признаков. ROC-AUC: 0.97, accuracy: 0.91, C-индекс = 0.616, accuracy: 0.91. Обе модели характеризуются высокой чувствительностью и специфичностью. </p> <p><bold>Заключение.</bold></p> <p>Полученные данные свидетельствуют о высоком потенциале радиомических маркеров, извлекаемых из Т1- и Т2-ВИ, для надежной дифференциальной диагностики доброкачественных и злокачественных образований яичников. </p></trans-abstract><trans-abstract xml:lang="zh"><p/></trans-abstract><kwd-group xml:lang="en"><kwd>radiomics, serous ovarian tumors, ovarian adenocarcinoma, ovarian cystadenoma, magnetic resonance imaging</kwd></kwd-group><kwd-group xml:lang="ru"><kwd>радиомика, серозные опухоли яичников, аденокарцинома яичника, цистаденома яичника, магнитно-резонансная томография</kwd></kwd-group><funding-group/></article-meta></front><body></body><back><ref-list><ref id="B1"><label>1.</label><citation-alternatives><mixed-citation xml:lang="en">1. https://www.who.int/ ( date of access 28.05.2025)</mixed-citation><mixed-citation xml:lang="ru">1. https://www.who.int/ (дата обращения 28.05.2025)</mixed-citation></citation-alternatives></ref><ref id="B2"><label>2.</label><mixed-citation>2. Bhatia, S., et al. (2019). "Imaging of ovarian tumors: A review." Journal of Clinical Imaging Science, 9, 1-10.</mixed-citation></ref><ref id="B3"><label>3.</label><citation-alternatives><mixed-citation xml:lang="en">3. B.P. Olimov, O.N. Streltsova, I.V. Panichenko. et al. IMAGING METHODS OF DIAGNOSIS OF UTERINE ADNEXAL TUMORS// Oncogynecology. 2018. №4.</mixed-citation><mixed-citation xml:lang="ru">3. Олимов Б. П. Лучевые методы диагностики опухолей придатков матки // Онкогинекология. 2018. №4. С. 39-40.</mixed-citation></citation-alternatives></ref><ref id="B4"><label>4.</label><mixed-citation>4. Vang R, Shih IeM, Kurman RJ. Ovarian low-grade and high-grade serous carcinoma: pathogenesis, clinicopathologic and molecular biologic features, and diagnostic problems. Adv Anat Pathol. 2009 Sep;16(5):267-82. doi: 10.1097/PAP.0b013e3181b4fffa. PMID: 19700937; PMCID: PMC2745605.</mixed-citation></ref><ref id="B5"><label>5.</label><mixed-citation>5. Arezzo, F.; Loizzi, V.; La Forgia, D.; Moschetta, M.; Tagliafico, A.S.; Cataldo, V.; Kawosha, A.A.; Venerito, V.; Cazzato, G.; Ingravallo, G.; et al. Radiomics Analysis in Ovarian Cancer: A Narrative Review. Appl. Sci. 2021, 11, 7833. https://doi.org/10.3390/app11177833</mixed-citation></ref><ref id="B6"><label>6.</label><mixed-citation>6. Huang ML, Ren J, Jin ZY, Liu XY, He YL, Li Y, Xue HD. A systematic review and meta-analysis of CT and MRI radiomics in ovarian cancer: methodological issues and clinical utility. Insights Imaging. 2023 Jul 3;14(1):117. doi: 10.1186/s13244-023-01464-z. PMID: 37395888; PMCID: PMC10317928.</mixed-citation></ref><ref id="B7"><label>7.</label><mixed-citation>7. Aerts, H. J. W. L., et al. (2014). "Decoding tumor phenotype by noninvasive imaging using a quantitative radiomics approach." Nature Communications, 5, 4006.</mixed-citation></ref><ref id="B8"><label>8.</label><mixed-citation>8. Lambin, P., et al. (2017). "Radiomics: The bridge between medical imaging and personalized medicine." Nature Reviews Clinical Oncology, 14(12), 749-762.</mixed-citation></ref><ref id="B9"><label>9.</label><mixed-citation>9. Kumar, V., et al. (2019). "Radiomics: The future of imaging in oncology." Nature Reviews Clinical Oncology, 16(12), 748-762.</mixed-citation></ref><ref id="B10"><label>10.</label><citation-alternatives><mixed-citation xml:lang="en">10. Xu Y, Luo HJ, Ren J, Guo LM, Niu J, Song X. Diffusion-weighted imaging-based radiomics in epithelial ovarian tumors: Assessment of histologic subtype. Front Oncol. 2022 Dec 5;12:978123. doi: 10.3389/fonc.2022.978123. PMID: 36544703; PMCID: PMC9762272.</mixed-citation><mixed-citation xml:lang="ru">10. Xu Y, Luo HJ, Ren J. et al. Diffusion-weighted imaging-based radiomics in epithelial ovarian tumors: Assessment of histologic subtype. Front Oncol. 2022 Dec 5;12:978123. doi: 10.3389/fonc.2022.978123. PMID: 36544703; PMCID: PMC9762272.</mixed-citation></citation-alternatives></ref><ref id="B11"><label>11.</label><mixed-citation>11. Wei, M., Zhang, Y., Bai, G. et al. T2-weighted MRI-based radiomics for discriminating between benign and borderline epithelial ovarian tumors: a multicenter study. Insights Imaging 13, 130 (2022). https://doi.org/10.1186/s13244-022-01264-x</mixed-citation></ref><ref id="B12"><label>12.</label><citation-alternatives><mixed-citation xml:lang="en">12. Liu X, Wang T, Zhang G, Hua K, Jiang H, Duan S, Jin J, Zhang H. Two-dimensional and three-dimensional T2 weighted imaging-based radiomic signatures for the preoperative discrimination of ovarian borderline tumors and malignant tumors. J Ovarian Res. 2022 Feb 3;15(1):22. doi: 10.1186/s13048-022-00943-z. PMID: 35115022; PMCID: PMC8815217.</mixed-citation><mixed-citation xml:lang="ru">12. Liu X, Wang T, Zhang G. et al. Two-dimensional and three-dimensional T2 weighted imaging-based radiomic signatures for the preoperative discrimination of ovarian borderline tumors and malignant tumors. J Ovarian Res. 2022 Feb 3;15(1):22. doi: 10.1186/s13048-022-00943-z. PMID: 35115022; PMCID: PMC8815217.</mixed-citation></citation-alternatives></ref><ref id="B13"><label>13.</label><citation-alternatives><mixed-citation xml:lang="en">13. Ye R, Weng S, Li Y, Yan C, Chen J, Zhu Y, Wen L. Texture Analysis of Three-Dimensional MRI Images May Differentiate Borderline and Malignant Epithelial Ovarian Tumors. Korean J Radiol. 2021 Jan;22(1):106-117. doi: 10.3348/kjr.2020.0121. Epub 2020 Sep 10. PMID: 32932563; PMCID: PMC7772386.</mixed-citation><mixed-citation xml:lang="ru">13. Ye R, Weng S, Li Y. et al. Texture Analysis of Three-Dimensional MRI Images May Differentiate Borderline and Malignant Epithelial Ovarian Tumors. Korean J Radiol. 2021 Jan;22(1):106-117. doi: 10.3348/kjr.2020.0121. Epub 2020 Sep 10. PMID: 32932563; PMCID: PMC7772386.</mixed-citation></citation-alternatives></ref><ref id="B14"><label>14.</label><mixed-citation>14. Song, Xl., Ren, JL., Zhao, D. et al. Radiomics derived from dynamic contrast-enhanced MRI pharmacokinetic protocol features: the value of precision diagnosis ovarian neoplasms. Eur Radiol 31, 368–378 (2021). https://doi.org/10.1007/s00330-020-07112-0</mixed-citation></ref></ref-list></back></article>
