<?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="review-article" dtd-version="1.2" xml:lang="en"><front><journal-meta><journal-id journal-id-type="publisher-id">N.N. Priorov Journal of Traumatology and Orthopedics</journal-id><journal-title-group><journal-title xml:lang="en">N.N. Priorov Journal of Traumatology and Orthopedics</journal-title><trans-title-group xml:lang="ru"><trans-title>Вестник травматологии и ортопедии им. Н.Н. Приорова</trans-title></trans-title-group></journal-title-group><issn publication-format="print">0869-8678</issn><issn publication-format="electronic">2658-6738</issn><publisher><publisher-name xml:lang="en">Eco-Vector</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="publisher-id">637087</article-id><article-id pub-id-type="doi">10.17816/vto637087</article-id><article-categories><subj-group subj-group-type="toc-heading" xml:lang="en"><subject>SCIENTIFIC REVIEWS</subject></subj-group><subj-group subj-group-type="toc-heading" xml:lang="ru"><subject>Научные обзоры</subject></subj-group><subj-group subj-group-type="article-type"><subject>Review Article</subject></subj-group></article-categories><title-group><article-title xml:lang="en">The influence of CNN architecture, image size and quality to object detection model on histological specimens</article-title><trans-title-group xml:lang="ru"><trans-title>Влияние архитектуры CNN, размера и качества изображений на модель, созданную для обнаружения объектов на гистологических препаратах</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-0829-9188</contrib-id><contrib-id contrib-id-type="spin">5380-3194</contrib-id><name-alternatives><name xml:lang="en"><surname>Fedosova</surname><given-names>Nina V.</given-names></name><name xml:lang="ru"><surname>Федосова</surname><given-names>Нина Вениаминовна</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MS</p></bio><email>hard_sign@mail.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-7920-0552</contrib-id><contrib-id contrib-id-type="spin">3367-2493</contrib-id><name-alternatives><name xml:lang="en"><surname>Berchenko</surname><given-names>Gennadiy N.</given-names></name><name xml:lang="ru"><surname>Берченко</surname><given-names>Геннадий Николаевич</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><email>berchenko@cito-bone.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-0778-5109</contrib-id><name-alternatives><name xml:lang="en"><surname>Shugaeva</surname><given-names>Olga B.</given-names></name><name xml:lang="ru"><surname>Шугаева</surname><given-names>Ольга Борисовна</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><email>Olga.schugaeva2013@yandex.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0003-5442-5055</contrib-id><contrib-id contrib-id-type="spin">5981-4084</contrib-id><name-alternatives><name xml:lang="en"><surname>Mashoshin</surname><given-names>Dmitriy V.</given-names></name><name xml:lang="ru"><surname>Машошин</surname><given-names>Дмитрий Викторович</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><email>dima_mash@mail.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0002-0699-1370</contrib-id><name-alternatives><name xml:lang="en"><surname>Kochan</surname><given-names>Mikhail G.</given-names></name><name xml:lang="ru"><surname>Кочан</surname><given-names>Михаил Геннадьевич</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><email>mk_system@mail.ru</email><xref ref-type="aff" rid="aff1"/></contrib></contrib-group><aff-alternatives id="aff1"><aff><institution xml:lang="en">Priorov National Medical Research Center of Traumatology and Orthopedics</institution></aff><aff><institution xml:lang="ru">Национальный медицинский исследовательский центр травматологии и ортопедии им. Н.Н. Приорова</institution></aff></aff-alternatives><pub-date date-type="preprint" iso-8601-date="2024-11-05" publication-format="electronic"><day>05</day><month>11</month><year>2024</year></pub-date><pub-date date-type="pub" iso-8601-date="2024-12-25" publication-format="electronic"><day>25</day><month>12</month><year>2024</year></pub-date><volume>31</volume><issue>4</issue><issue-title xml:lang="en"/><issue-title xml:lang="ru"/><fpage>751</fpage><lpage>758</lpage><history><date date-type="received" iso-8601-date="2024-10-15"><day>15</day><month>10</month><year>2024</year></date><date date-type="accepted" iso-8601-date="2024-10-25"><day>25</day><month>10</month><year>2024</year></date></history><permissions><copyright-statement xml:lang="en">Copyright ©; 2024, Eco-Vector</copyright-statement><copyright-statement xml:lang="ru">Copyright ©; 2024, Эко-Вектор</copyright-statement><copyright-year>2024</copyright-year><copyright-holder xml:lang="en">Eco-Vector</copyright-holder><copyright-holder xml:lang="ru">Эко-Вектор</copyright-holder><ali:free_to_read xmlns:ali="http://www.niso.org/schemas/ali/1.0/" start_date="2025-12-25"/><license><ali:license_ref xmlns:ali="http://www.niso.org/schemas/ali/1.0/">https://eco-vector.com/for_authors.php#07</ali:license_ref></license></permissions><self-uri xlink:href="https://journals.eco-vector.com/0869-8678/article/view/637087">https://journals.eco-vector.com/0869-8678/article/view/637087</self-uri><abstract xml:lang="en"><p>Improving convolutional neural network (CNN) quality for object search in histology scans is a long-standing problem that essentially involves selecting the best CNN architecture and creating a high-quality dataset. The efficacy of object detection algorithms is determined by numerous factors, including image quality, image size, and the search object. The primary aim of this study was to identify published studies on the impact of various image characteristics in a training sample and CNN architecture on the quality of a created model. Literature published in the last 5 years was reviewed, which addressed data pre-processing, methodology, requirements to images included in datasets, image preparation for CNN model development, and architecture selection. At the time of the study, there were no requirements to image size, and there was no data on the ratio of object size to image size for the best model performance. Moreover, the selection of neural network architecture is lacking in transparency and algorithmization. In the majority of cases, researchers recommend architectures that they have developed or used themselves, without explaining the reasons and selection criteria or comparing them to alternative options. All these factors significantly complicate the development of CNN models for medical image processing. This paper presents a brief overview of publications that address image preparation for datasets, as well as a potential approach to CNN architecture selection.</p></abstract><trans-abstract xml:lang="ru"><p>Проблема повышения качества свёрточной нейронной сети (англ. convolutional neural network, CNN) в случае поиска объектов на гистологических сканах возникла достаточно давно и сводится прежде всего к выбору оптимальной архитектуры CNN и подготовке датасета надлежащего качества. На работу алгоритмов обнаружения объектов влияют многие факторы, включая качество изображения, размер изображения и самого объекта поиска. Основной целью проведённой работы являлся поиск существующих исследований, демонстрирующих влияние различных характеристик изображений в обучающей выборке и выбора архитектуры CNN на качество создаваемой модели. В качестве исходных материалов для работы была проанализирована литература за последние 5 лет, посвящённая вопросам предварительной обработки данных, методологиям, требованиям к изображениям, включаемым в датасеты, подготовке изображений для создания моделей CNN и вопросам выбора архитектуры. На момент проведения исследования не сформулировано требований к размерам изображений и отсутствуют данные о соотношении размеров объектов и размеров изображения для оптимизации работы модели. Кроме того, задача выбора архитектуры нейронной сети не является прозрачной и алгоритмизированной. В большинстве случаев исследователи рекомендуют те архитектуры, которые сами разрабатывали или использовали, не объясняя ни причины, ни критерии выбора, не сравнивая с альтернативными вариантами. Всё это существенно затрудняет процесс разработки модели CNN для обработки медицинских изображений. В данном обзорном исследовании приводится краткий обзор исследований, посвящённых подготовке изображений для датасета и возможного подхода к выбору архитектуры CNN.</p></trans-abstract><kwd-group xml:lang="en"><kwd>neural network</kwd><kwd>mathematical model</kwd><kwd>artificial intelligence</kwd><kwd>machine learning</kwd></kwd-group><kwd-group xml:lang="ru"><kwd>нейронная сеть</kwd><kwd>математическая модель</kwd><kwd>искусственный интеллект</kwd><kwd>машинное обучение</kwd></kwd-group><funding-group><award-group><funding-source><institution-wrap><institution xml:lang="ru">Правительство Российской Федерации</institution></institution-wrap><institution-wrap><institution xml:lang="en">Government of the Russian Federation</institution></institution-wrap></funding-source><award-id>124040100041-5</award-id></award-group></funding-group></article-meta></front><body></body><back><ref-list><ref id="B1"><label>1.</label><citation-alternatives><mixed-citation xml:lang="en">Hort M, Chen Z, Zhang JM, Harman M, Sarro F. Bias Mitigation for Machine Learning Classifiers: A Comprehensive Survey. ACM Journal on Responsible Computing. 2024;1(2):1–52. doi: 10.1145/3631326</mixed-citation><mixed-citation xml:lang="ru">Hort M., Chen Z., Zhang J.M., Harman M., Sarro F. Bias Mitigation for Machine Learning Classifiers: A Comprehensive Survey // ACM Journal on Responsible Computing. 2024. Vol. 1, № 2. Р. 1–52. doi: 10.1145/3631326</mixed-citation></citation-alternatives></ref><ref id="B2"><label>2.</label><citation-alternatives><mixed-citation xml:lang="en">Kamiran F, Toon C. Data preprocessing techniques for classification without discrimination. Knowledge and Information Systems. 2012;33(1):1–33. doi: 10.1007/s10115-011-0463-8</mixed-citation><mixed-citation xml:lang="ru">Kamiran F., Toon C. Data preprocessing techniques for classification without discrimination // Knowledge and Information Systems. 2012. Vol. 33, № 1. Р. 1–33. doi: 10.1007/s10115-011-0463-8</mixed-citation></citation-alternatives></ref><ref id="B3"><label>3.</label><citation-alternatives><mixed-citation xml:lang="en">Li T, Chen K-S, Jin M. Analysis and simulation on imaging performance of backward and forward bistatic synthetic aperture radar. Remote Sensing. 2018;10(11):1676. doi: 10.3390/rs10111676</mixed-citation><mixed-citation xml:lang="ru">Li T., Chen K.-S., Jin M. Analysis and simulation on imaging performance of backward and forward bistatic synthetic aperture radar // Remote Sensing. 2018. Vol. 10, № 11. P. 1676. doi: 10.3390/rs10111676</mixed-citation></citation-alternatives></ref><ref id="B4"><label>4.</label><citation-alternatives><mixed-citation xml:lang="en">Rius A, Cardellach E, Fabra F, et al. Feasibility of GNSS-R ice sheet altimetry in greenland using TDS-1. Remote Sensing. 2017;9(7):742. doi: 10.3390/rs9070742</mixed-citation><mixed-citation xml:lang="ru">Rius A., Cardellach E., Fabra F., et al. Feasibility of GNSS-R ice sheet altimetry in greenland using TDS-1 // Remote Sensing. 2017. Vol. 9, № 7. P. 742. doi: 10.3390/rs9070742</mixed-citation></citation-alternatives></ref><ref id="B5"><label>5.</label><citation-alternatives><mixed-citation xml:lang="en">Yan J, Liu X, Wang X. Object detection in videos with tubelet proposal networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition. 2017. Р. 727–735. doi: 10.1109/cvpr.2017.101</mixed-citation><mixed-citation xml:lang="ru">Yan J., Liu X., Wang X. Object detection in videos with tubelet proposal networks. В кн.: Proceedings of the IEEE conference on computer vision and pattern recognition. 2017. Р. 727–735. doi: 10.1109/cvpr.2017.101</mixed-citation></citation-alternatives></ref><ref id="B6"><label>6.</label><citation-alternatives><mixed-citation xml:lang="en">Kendall A, Gal Y, Cipolla R. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics. In: Proceedings of the IEEE conference on computer vision and pattern recognition. 2018. Р. 7482–7491. doi: 10.1109/cvpr.2018.00781</mixed-citation><mixed-citation xml:lang="ru">Kendall A., Gal Y., Cipolla R. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics. В кн.: Proceedings of the IEEE conference on computer vision and pattern recognition. 2018. Р. 7482–7491. doi: 10.1109/cvpr.2018.00781</mixed-citation></citation-alternatives></ref><ref id="B7"><label>7.</label><citation-alternatives><mixed-citation xml:lang="en">Chen W, Daneau S, Mannan F, Heide F. Steady-state nonline- of-sight imaging. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2019. Р. 6790–6799. doi: 10.1109/cvpr.2019.00695</mixed-citation><mixed-citation xml:lang="ru">Chen W., Daneau S., Mannan F., Heide F. Steady-state nonline- of-sight imaging. В кн.: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2019. Р. 6790–6799. doi: 10.1109/cvpr.2019.00695</mixed-citation></citation-alternatives></ref><ref id="B8"><label>8.</label><citation-alternatives><mixed-citation xml:lang="en">Kaneyasu H, Etter SB, Sakai T, Sigrist M. Evolution of the filamentary 3-Kelvin phase in Pb-Ru-Sr2RuO4 Josephson junctions. Physical Review B. 2015;92(13):134515. doi: 10.1103/physrevb.92.134515</mixed-citation><mixed-citation xml:lang="ru">Kaneyasu H., Etter S.B., Sakai T., Sigrist M. Evolution of the filamentary 3-Kelvin phase in Pb-Ru-Sr2RuO4 Josephson junctions // Physical Review B. 2015. Vol. 92, № 13. P. 134515. doi: 10.1103/physrevb.92.134515</mixed-citation></citation-alternatives></ref><ref id="B9"><label>9.</label><citation-alternatives><mixed-citation xml:lang="en">Ren S, He K, Girshick R, Sun J. Faster R-CNN: Towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2015;28. doi: 10.1109/tpami.2016.2577031</mixed-citation><mixed-citation xml:lang="ru">Ren S., He K., Girshick R., Sun J. Faster R-CNN: Towards real-time object detection with region proposal networks // IEEE Transactions on Pattern Analysis and Machine Intelligence. 2015. Vol. 28. doi: 10.1109/tpami.2016.2577031</mixed-citation></citation-alternatives></ref><ref id="B10"><label>10.</label><citation-alternatives><mixed-citation xml:lang="en">Liu W, Anguelov D, Erhan D, et al. SSD: Single shot multibox detector. In: Computer Vision–ECCV. Part I. Springer; 2016. Р. 21–37. doi: 10.1007/978-3-319-46448-0_2</mixed-citation><mixed-citation xml:lang="ru">Liu W., Anguelov D., Erhan D., et al. SSD: Single shot multibox detector. В кн.: Computer Vision–ECCV. Part I. Springer, 2016. Р. 21–37. doi: 10.1007/978-3-319-46448-0_2</mixed-citation></citation-alternatives></ref><ref id="B11"><label>11.</label><citation-alternatives><mixed-citation xml:lang="en">Redmon J, Farhadi A. YOLOv3: An incremental improvement. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2017. doi: 10.1109/cvpr.2017.690</mixed-citation><mixed-citation xml:lang="ru">Redmon J., Farhadi A. YOLOv3: An incremental improvement. В кн.: IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2017. doi: 10.1109/cvpr.2017.690</mixed-citation></citation-alternatives></ref><ref id="B12"><label>12.</label><citation-alternatives><mixed-citation xml:lang="en">Zhang S, Zhu X, Lei Z, et al. S3FD: Single shot scale-invariant face detector. In: IEEE international conference on computer vision. 2017. Р. 192–201. doi: 10.1109/iccv.2017.30</mixed-citation><mixed-citation xml:lang="ru">Zhang S., Zhu X., Lei Z., et al. S3FD: Single shot scale-invariant face detector. В кн.: IEEE international conference on computer vision. 2017. Р. 192–201. doi: 10.1109/iccv.2017.30</mixed-citation></citation-alternatives></ref><ref id="B13"><label>13.</label><citation-alternatives><mixed-citation xml:lang="en">Kumar A, Mital U, Gajera A, Varanasi S, Patra D. Empirical Study of the Impact of Image Quality, Object size, and Occlusion to Object Detection [Internet]. EasyChair Preprint 9786. 2023. Available from: https://easychair.org/publications/preprint/Wf1V</mixed-citation><mixed-citation xml:lang="ru">Kumar A., Mital U., Gajera A., Varanasi S., Patra D. Empirical Study of the Impact of Image Quality, Object size, and Occlusion to Object Detection [Internet]. EasyChair Preprint 9786. 2023. Режим доступа: https://easychair.org/publications/preprint/Wf1V</mixed-citation></citation-alternatives></ref><ref id="B14"><label>14.</label><citation-alternatives><mixed-citation xml:lang="en">Buyukkinaci M. Fruit images for object detection [Internet]. Available from: https://www.kaggle.com/datasets/mbkinaci/fruit-images-for-objectdetection</mixed-citation><mixed-citation xml:lang="ru">Buyukkinaci M. Fruit images for object detection [Internet]. Режим доступа: https://www.kaggle.com/datasets/mbkinaci/fruit-images-for-objectdetection</mixed-citation></citation-alternatives></ref><ref id="B15"><label>15.</label><citation-alternatives><mixed-citation xml:lang="en">Hao Y, Pei H, Lyu Y, et al. Understanding the Impact of Image Quality and Distance of Objects to Object Detection Performance. NYU Multimedia and Visual Computing Lab. 2022. doi: 10.48550/arXiv.2209.08237</mixed-citation><mixed-citation xml:lang="ru">Hao Y., Pei H., Lyu Y., et al. Understanding the Impact of Image Quality and Distance of Objects to Object Detection Performance // NYU Multimedia and Visual Computing Lab. 2022. doi: 10.48550/arXiv.2209.08237</mixed-citation></citation-alternatives></ref><ref id="B16"><label>16.</label><citation-alternatives><mixed-citation xml:lang="en">Seker E, Talburt JR, Greer ML. Preprocessing to Address Bias in Healthcare Data. Studies in Health Technology and Informatics. 2022;294:327–331. doi: 10.3233/shti220468</mixed-citation><mixed-citation xml:lang="ru">Seker E., Talburt J.R., Greer M.L. Preprocessing to Address Bias in Healthcare Data // Studies in Health Technology and Informatics. 2022. Vol. 294. Р. 327–331. doi: 10.3233/shti220468</mixed-citation></citation-alternatives></ref><ref id="B17"><label>17.</label><citation-alternatives><mixed-citation xml:lang="en">Celis LE, Keswani V, Vishnoi N. Data preprocessing to mitigate bias: A maximum entropy-based approach. In: International Conference on Machine Learning, PMLR. 2020. Р. 1349–1359. doi: 10.1201/9781003055129-6</mixed-citation><mixed-citation xml:lang="ru">Celis L.E., Keswani V., Vishnoi N. Data preprocessing to mitigate bias: A maximum entropy-based approach. В кн.: International Conference on Machine Learning, PMLR. 2020. Р. 1349–1359. doi: 10.1201/9781003055129-6</mixed-citation></citation-alternatives></ref></ref-list></back></article>
