<?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">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">678373</article-id><article-id pub-id-type="doi">10.17816/DD678373</article-id><article-id pub-id-type="edn">QSANCA</article-id><article-categories><subj-group subj-group-type="toc-heading" xml:lang="en"><subject>Systematic reviews</subject></subj-group><subj-group subj-group-type="toc-heading" xml:lang="ru"><subject>Систематические обзоры</subject></subj-group><subj-group subj-group-type="toc-heading" xml:lang="zh"><subject>系统评价</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">Application of large language models in radiological diagnostics: a scoping review</article-title><trans-title-group xml:lang="ru"><trans-title>Применение больших языковых моделей в лучевой диагностике: обзор предметного поля</trans-title></trans-title-group><trans-title-group xml:lang="zh"><trans-title>大语言模型在放射诊断中的应用：范围综述</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-5283-5961</contrib-id><contrib-id contrib-id-type="spin">4458-5608</contrib-id><name-alternatives><name xml:lang="en"><surname>Vasilev</surname><given-names>Yuriy A.</given-names></name><name xml:lang="ru"><surname>Васильев</surname><given-names>Юрий Александрович</given-names></name><name xml:lang="zh"><surname>Vasilev</surname><given-names>Yuriy A.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD, Cand. Sci. (Medicine)</p></bio><bio xml:lang="ru"><p>канд. мед. наук</p></bio><bio xml:lang="zh"><p>MD, Cand. Sci. (Medicine)</p></bio><email>npcmr@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-9661-0254</contrib-id><contrib-id contrib-id-type="spin">8592-0558</contrib-id><name-alternatives><name xml:lang="en"><surname>Reshetnikov</surname><given-names>Roman V.</given-names></name><name xml:lang="ru"><surname>Решетников</surname><given-names>Роман Владимирович</given-names></name><name xml:lang="zh"><surname>Reshetnikov</surname><given-names>Roman V.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>Cand. Sci. (Physics and Mathematics)</p></bio><bio xml:lang="ru"><p>канд. физ.-мат. наук</p></bio><bio xml:lang="zh"><p>Cand. Sci. (Physics and Mathematics)</p></bio><email>ReshetnikovRV1@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0001-8886-3684</contrib-id><contrib-id contrib-id-type="spin">6135-4872</contrib-id><name-alternatives><name xml:lang="en"><surname>Nanova</surname><given-names>Olga G.</given-names></name><name xml:lang="ru"><surname>Нанова</surname><given-names>Ольга Геннадьевна</given-names></name><name xml:lang="zh"><surname>Nanova</surname><given-names>Olga G.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>Cand. Sci. (Biology)</p></bio><bio xml:lang="ru"><p>канд. биол. наук</p></bio><bio xml:lang="zh"><p>Cand. Sci. (Biology)</p></bio><email>nanova@mail.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-2990-7736</contrib-id><contrib-id contrib-id-type="spin">3602-7120</contrib-id><name-alternatives><name xml:lang="en"><surname>Vladzymyrskyy</surname><given-names>Anton V.</given-names></name><name xml:lang="ru"><surname>Владзимирский</surname><given-names>Антон Вячеславович</given-names></name><name xml:lang="zh"><surname>Vladzymyrskyy</surname><given-names>Anton V.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD, Dr. Sci. (Medicine)</p></bio><bio xml:lang="ru"><p>д-р мед. наук</p></bio><bio xml:lang="zh"><p>MD, Dr. Sci. (Medicine)</p></bio><email>VladzimirskijAV@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0001-7786-0349</contrib-id><contrib-id contrib-id-type="spin">3160-8062</contrib-id><name-alternatives><name xml:lang="en"><surname>Arzamasov</surname><given-names>Kirill M.</given-names></name><name xml:lang="ru"><surname>Арзамасов</surname><given-names>Кирилл Михайлович</given-names></name><name xml:lang="zh"><surname>Arzamasov</surname><given-names>Kirill M.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD, Dr. Sci. (Medicine)</p></bio><bio xml:lang="ru"><p>д-р мед. наук</p></bio><bio xml:lang="zh"><p>MD, Dr. Sci. (Medicine)</p></bio><email>ArzamasovKM@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-0245-4431</contrib-id><contrib-id contrib-id-type="spin">8948-6152</contrib-id><name-alternatives><name xml:lang="en"><surname>Omelyanskaya</surname><given-names>Olga V.</given-names></name><name xml:lang="ru"><surname>Омелянская</surname><given-names>Ольга Васильевна</given-names></name><name xml:lang="zh"><surname>Omelyanskaya</surname><given-names>Olga V.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><email>o.omelyanskaya@npcmr.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-0166-3768</contrib-id><contrib-id contrib-id-type="spin">5789-0319</contrib-id><name-alternatives><name xml:lang="en"><surname>Kodenko</surname><given-names>Maria R.</given-names></name><name xml:lang="ru"><surname>Коденко</surname><given-names>Мария Романовна</given-names></name><name xml:lang="zh"><surname>Kodenko</surname><given-names>Maria R.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>Cand. Sci. (Engineering)</p></bio><bio xml:lang="ru"><p>канд. техн. наук</p></bio><bio xml:lang="zh"><p>Cand. Sci. (Engineering)</p></bio><email>KodenkoMR@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0007-3636-2889</contrib-id><contrib-id contrib-id-type="spin">2274-6428</contrib-id><name-alternatives><name xml:lang="en"><surname>Erizhokov</surname><given-names>Rustam A.</given-names></name><name xml:lang="ru"><surname>Ерижоков</surname><given-names>Рустам Арсеньевич</given-names></name><name xml:lang="zh"><surname>Erizhokov</surname><given-names>Rustam A.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD</p></bio><bio xml:lang="zh"><p>MD</p></bio><email>ErizhokovRA@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-0041-3281</contrib-id><contrib-id contrib-id-type="spin">5146-4355</contrib-id><name-alternatives><name xml:lang="en"><surname>Pamova</surname><given-names>Anastasia P.</given-names></name><name xml:lang="ru"><surname>Памова</surname><given-names>Анастасия Петровна</given-names></name><name xml:lang="zh"><surname>Pamova</surname><given-names>Anastasia P.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD, Cand. Sci. (Medicine)</p></bio><bio xml:lang="ru"><p>канд. мед. наук</p></bio><bio xml:lang="zh"><p>MD, Cand. Sci. (Medicine)</p></bio><email>PamovaAP@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0000-3990-6668</contrib-id><name-alternatives><name xml:lang="en"><surname>Seradzhi</surname><given-names>Seal R.</given-names></name><name xml:lang="ru"><surname>Сераджи</surname><given-names>Сеал Рахмануддин</given-names></name><name xml:lang="zh"><surname>Seradzhi</surname><given-names>Seal R.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><email>SeradzhiSR@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-2681-9378</contrib-id><contrib-id contrib-id-type="spin">3306-1387</contrib-id><name-alternatives><name xml:lang="en"><surname>Blokhin</surname><given-names>Ivan A.</given-names></name><name xml:lang="ru"><surname>Блохин</surname><given-names>Иван Андреевич</given-names></name><name xml:lang="zh"><surname>Blokhin</surname><given-names>Ivan A.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD, Cand. Sci. (Medicine)</p></bio><bio xml:lang="ru"><p>канд. мед. наук</p></bio><bio xml:lang="zh"><p>MD, Cand. Sci. (Medicine)</p></bio><email>BlokhinIA@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0001-5161-6540</contrib-id><contrib-id contrib-id-type="spin">3513-9531</contrib-id><name-alternatives><name xml:lang="en"><surname>Gonchar</surname><given-names>Anna P.</given-names></name><name xml:lang="ru"><surname>Гончар</surname><given-names>Анна Павловна</given-names></name><name xml:lang="zh"><surname>Gonchar</surname><given-names>Anna P.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD, Cand. Sci. (Medicine)</p></bio><bio xml:lang="ru"><p>канд. мед. наук</p></bio><bio xml:lang="zh"><p>MD, Cand. Sci. (Medicine)</p></bio><email>GoncharAP@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/><xref ref-type="aff" rid="aff2"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0003-1072-2202</contrib-id><contrib-id contrib-id-type="spin">4841-3234</contrib-id><name-alternatives><name xml:lang="en"><surname>Gelezhe</surname><given-names>Pavel B.</given-names></name><name xml:lang="ru"><surname>Гележе</surname><given-names>Павел Борисович</given-names></name><name xml:lang="zh"><surname>Gelezhe</surname><given-names>Pavel B.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD, Cand. Sci. (Medicine)</p></bio><bio xml:lang="ru"><p>канд. мед. наук</p></bio><bio xml:lang="zh"><p>MD, Cand. Sci. (Medicine)</p></bio><email>GelezhePB@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0001-7705-9754</contrib-id><contrib-id contrib-id-type="spin">6983-5991</contrib-id><name-alternatives><name xml:lang="en"><surname>Akhmedzyanova</surname><given-names>Dina A.</given-names></name><name xml:lang="ru"><surname>Ахмедзянова</surname><given-names>Дина Альфредовна</given-names></name><name xml:lang="zh"><surname>Akhmedzyanova</surname><given-names>Dina A.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD</p></bio><bio xml:lang="zh"><p>MD</p></bio><email>AkhmedzyanovaDA@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-8521-4045</contrib-id><contrib-id contrib-id-type="spin">3164-5518</contrib-id><name-alternatives><name xml:lang="en"><surname>Shumskaya</surname><given-names>Yuliya F.</given-names></name><name xml:lang="ru"><surname>Шумская</surname><given-names>Юлия Федоровна</given-names></name><name xml:lang="zh"><surname>Shumskaya</surname><given-names>Yuliya F.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD</p></bio><bio xml:lang="zh"><p>MD</p></bio><email>shumskayayf@zdrav.mos.ru</email><xref ref-type="aff" rid="aff1"/></contrib></contrib-group><aff-alternatives id="aff1"><aff><institution xml:lang="en">Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies</institution></aff><aff><institution xml:lang="ru">Научно-практический клинический центр диагностики и телемедицинских технологий</institution></aff><aff><institution xml:lang="zh">Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies</institution></aff></aff-alternatives><aff-alternatives id="aff2"><aff><institution xml:lang="en">Moscow City Hospital named after S.S. Yudin</institution></aff><aff><institution xml:lang="ru">Городская клиническая больница им. С.С. Юдина</institution></aff><aff><institution xml:lang="zh">Moscow City Hospital named after S.S. Yudin</institution></aff></aff-alternatives><pub-date date-type="preprint" iso-8601-date="2025-06-17" publication-format="electronic"><day>17</day><month>06</month><year>2025</year></pub-date><pub-date date-type="pub" iso-8601-date="2025-07-08" publication-format="electronic"><day>08</day><month>07</month><year>2025</year></pub-date><volume>6</volume><issue>2</issue><issue-title xml:lang="ru"/><fpage>268</fpage><lpage>285</lpage><history><date date-type="received" iso-8601-date="2025-05-06"><day>06</day><month>05</month><year>2025</year></date><date date-type="accepted" iso-8601-date="2025-06-12"><day>12</day><month>06</month><year>2025</year></date></history><permissions><copyright-statement xml:lang="en">Copyright ©; 2025, Eco-Vector</copyright-statement><copyright-statement xml:lang="ru">Copyright ©; 2025, Эко-вектор</copyright-statement><copyright-statement xml:lang="zh">Copyright ©; 2025, Eco-Vector</copyright-statement><copyright-year>2025</copyright-year><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/678373">https://jdigitaldiagnostics.com/DD/article/view/678373</self-uri><abstract xml:lang="en"><p><bold><italic>BACKGROUND: </italic></bold>Modern large language models show potential for application in radiological diagnostics across a wide range of routine tasks.</p> <p><bold><italic>AIM: </italic></bold>The work aimed to conduct a scoping review of the application of large language models in radiological diagnostics by analyzing possible use-case scenarios and assessing the methodological quality of relevant studies.</p> <p><bold><italic>METHODS: </italic></bold>Two search strategies were employed: a primary search (PubMed and eLibrary) targeting full-text publications with well-developed methodology, and a supplementary search (PubMed) aimed at broader coverage of large language model use cases in radiological diagnostics during 2023–2025. Extracted data included bibliometric characteristics, study objectives, use-case scenarios of large language models, nosological profiles, key methodological parameters, and both quantitative and qualitative indicators of diagnostic performance—for both the models and the specialists involved, including their number and experience. The quality was assessed using the modified QUADAS-CAD questionnaire.</p> <p><bold><italic>RESULTS: </italic></bold>The primary search yielded 9 studies for analysis; the supplementary search yielded 216. A total of 9 major use-case scenarios for large language models in radiology were identified. The most common among them was the rephrasing of radiology reports in order to improve their accessibility for patient understanding. Models predominantly used were GPT-4 and BERT, along with GPT-3.5, Llama 2, Med42, GPT-4V, and Gemini Pro. The large language model GPT-4 demonstrated high diagnostic accuracy in identifying brain tumors (73.0%), myocarditis (83.0%), and in making decisions on invasive procedures for acute coronary syndrome (86.0%). In turn, it demonstrated low diagnostic accuracy for nervous system disorders of various etiologies (50.0%) and for musculoskeletal diseases (43.0%). The BERT model exhibited high diagnostic accuracy in detecting pulmonary nodules (99.0%) and signs of intracranial hemorrhage (sensitivity and specificity: 97.0% and 90.0%, respectively), as well as in report classification (accuracy: 84.3%).</p> <p>Most articles (88.9%) carried a high risk of bias. The main reasons for this included small and imbalanced sample sizes, overlap between training and test datasets, and insufficiently precise preparation and description of reference standards.</p> <p><bold><italic>CONCLUSION: </italic></bold>The diagnostic performance of large language models varies significantly across articles. Their clinical implementation requires standardized, methodologically robust research, including larger and more balanced samples, optimization of the structure and volume of datasets, separation of training and testing samples, thorough preparation and description of reference standards, as well as the accumulation of empirical data for specific radiological tasks.</p></abstract><trans-abstract xml:lang="ru"><p><bold>Обоснование. </bold>Современные большие языковые модели обладают потенциалом использования в лучевой диагностике для решения широкого спектра рутинных задач.</p> <p><bold>Цель исследования. </bold>Провести обзор предметного поля применения больших языковых моделей в лучевой диагностике с анализом возможных сценариев их использования и оценкой качества методологии соответствующих исследований.</p> <p><bold>Методы. </bold>Провели два варианта поиска — первичный (PubMed и eLibrary), ориентированный на выявление полнотекстовых публикаций с максимально проработанной методологией, и дополнительный (PubMed), направленный на широкий охват сценариев применения больших языковых моделей в лучевой диагностике за период 2023–2025 гг. Извлекали библиометрические данные, формулировку исследовательской задачи, сценарий применения больших языковых моделей, нозологический профиль, ключевые методологические параметры, а также количественные и качественные показатели диагностической эффективности как моделей, так и участвующих специалистов, включая их число и опыт. Качество исследований оценивали с использованием модифицированного опросника QUADAS-CAD.</p> <p><bold>Результаты. </bold>При первичном поиске для анализа отобрано 9 публикаций, при дополнительном — 216. Найдено 9 основных сценариев применения больших языковых моделей в лучевой диагностике. Наиболее распространёнными из них было переформулирование рентгенологических заключений с целью повышения их доступности восприятия пациентами. Преимущественно использовали модели GPT-4 и BERT, а также GPT-3.5, Llama 2, Med42, GPT-4V и Gemini Pro. Большая языковая модель GPT-4 продемонстрировала высокую точность при диагностике опухолей головного мозга (73,0%), миокардитов (83,0%), а также в случае принятия решений о проведении инвазивной процедуры при остром коронарном синдроме (86,0%). В свою очередь, она продемонстрировала низкую диагностическую точность в отношении патологий нервной системы различной этиологии (50,0%) и заболеваний опорно-двигательной системы (43,0%). Модель BERT показала высокую диагностическую точность в задачах детекции лёгочных узелков (99,0%) и признаков внутричерепного кровоизлияния (чувствительность и специфичность — 97,0 и 90,0% соответственно), а также при классификации заключений (точность 84,3%).</p> <p>Большинство работ (88,9%) содержат вероятность систематической ошибки. Основные причины этого: маленький объём и несбалансированность выборок, пересечение обучающих и тестовых наборов данных, недостаточно аккуратная подготовка и описание референсных стандартов.</p> <p><bold>Заключение. </bold>Показатели диагностической точности больших языковых моделей сильно варьируют между разными исследованиями. Для их внедрения в клиническую практику необходимо проведение стандартизированных и методологически качественных исследований, включающих увеличение объёма и сбалансированности выборок, оптимизацию структуры и объёма наборов данных, формирование неперекрывающихся обучающих и тестовых выборок, тщательную подготовку и описание референсных стандартов, а также накопление эмпирических данных по отдельным задачам лучевой диагностики.</p></trans-abstract><trans-abstract xml:lang="zh"><p>论证。现代大语言模型具备在放射诊断中应用于解决广泛常规任务的潜力。</p> <p>目的：综述大语言模型在放射诊断中的应用范围，分析其使用场景，并评估相关研究的方法学质量。</p> <p>方法。开展两轮文献检索：初步检索（PubMed和eLibrary）聚焦于具备详实方法学的全文研究，补充检索（PubMed）旨在广泛覆盖2023–2025年间大语言模型在放射诊断中应用的各种情境。提取了书目信息、研究任务的表述、大语言模型的应用场景、疾病谱、关键方法学参数，以及诊断效能的定量与定性指标，涵盖模型本身及参与专家，包括其人数与经验。采用改良版QUADAS-CAD问卷对研究质量进行评估。</p> <p>结果。初步检索纳入9项研究，补充检索纳入216项。共识别出在放射诊断中应用大语言模型的9种主要场景。其中最常见的是为提升患者理解而对放射学报告进行改写。最常使用的模型包括GPT-4和BERT，以及GPT-3.5、Llama 2、Med42、GPT-4V和Gemini Pro。大语言模型GPT-4在脑肿瘤（73.0%）、心肌炎（83.0%）以及急性冠状动脉综合征中介入治疗决策（86.0%）方面表现出较高的诊断准确性。但在诊断不同病因的神经系统疾病（50.0%）和肌肉骨骼疾病（43.0%）方面准确性较低。BERT模型在肺结节检测（99.0%）和颅内出血征象识别（灵敏度97.0%、特异度90.0%）方面表现优异，在放射学报告分类中准确率为84.3%。</p> <p>大多数研究（88.9%）存在系统性偏倚的可能。其主要原因包括：样本量小且分布不均、训练集与测试集重叠、参考标准准备和描述不够严谨。</p> <p>结论。大语言模型的诊断准确性在不同研究间差异显著。其进入临床实践前，亟需开展标准化且方法学严谨的研究，包括扩大并平衡样本量、优化数据集结构与规模、明确划分训练集与测试集、严谨制定和描述参考标准，并针对特定放射诊断任务积累实证数据。</p></trans-abstract><kwd-group xml:lang="en"><kwd>artificial intelligence</kwd><kwd>large language models</kwd><kwd>radiological diagnostics</kwd><kwd>radiology report</kwd><kwd>systematic review</kwd></kwd-group><kwd-group xml:lang="ru"><kwd>искусственный интеллект</kwd><kwd>большие языковые модели</kwd><kwd>лучевая диагностика</kwd><kwd>рентгенологический протокол</kwd><kwd>систематический обзор</kwd></kwd-group><kwd-group xml:lang="zh"><kwd>人工智能</kwd><kwd>大语言模型</kwd><kwd>放射诊断</kwd><kwd>放射学报告</kwd><kwd>系统综述</kwd></kwd-group><funding-group><award-group><funding-source><institution-wrap><institution xml:lang="en">Moscow City Health Department</institution></institution-wrap><institution-wrap><institution xml:lang="ru">Департамент здравоохранения города Москвы</institution></institution-wrap><institution-wrap><institution xml:lang="zh">Moscow City Health Department</institution></institution-wrap></funding-source><award-id>656-ПП</award-id></award-group></funding-group></article-meta></front><body></body><back><ref-list><ref id="B1"><label>1.</label><mixed-citation>Cherif H, Moussa C, Missaoui AM, et al. Appraisal of ChatGPT’s aptitude for medical education: comparative analysis with third-year medical students in a pulmonology examination. JMIR Medical Education. 2024;10:e52818. doi: 10.2196/52818 EDN: OFMTDE</mixed-citation></ref><ref id="B2"><label>2.</label><mixed-citation>Kim W, Kim BC, Yeom HG. Performance of large language models on the Korean Dental licensing examination: a comparative study. International Dental Journal. 2025;75(1):176–184. doi: 10.1016/j.identj.2024.09.002 EDN: JDFMDL</mixed-citation></ref><ref id="B3"><label>3.</label><mixed-citation>Busch F, Hoffmann L, dos Santos DP, et al. Large language models for structured reporting in radiology: past, present, and future. European Radiology. 2024;35(5):2589–2602. doi: 10.1007/s00330-024-11107-6 EDN: PNFKNR</mixed-citation></ref><ref id="B4"><label>4.</label><mixed-citation>Lecler A, Duron L, Soyer P. Revolutionizing radiology with GPT-based models: Current applications, future possibilities and limitations of ChatGPT. Diagnostic and Interventional Imaging. 2023;104(6):269–274. doi: 10.1016/j.diii.2023.02.003EDN: FGMMTY</mixed-citation></ref><ref id="B5"><label>5.</label><mixed-citation>Tricco AC, Lillie E, Zarin W, et al. PRISMA Extension for Scoping Reviews (PRISMA-ScR): Checklist and Explanation. Annals of Internal Medicine. 2018;169(7):467–473. doi: 10.7326/M18-0850</mixed-citation></ref><ref id="B6"><label>6.</label><mixed-citation>Vasilev YuA, Vladzymyrskyy AV, Omelyanskaya OV, et al. Methodological recommendations for preparing a systematic review. Moscow: Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies; 2023. (In Russ.) EDN: XKXHDA</mixed-citation></ref><ref id="B7"><label>7.</label><mixed-citation>Kodenko MR, Vasilev YA, Vladzymyrskyy AV, et al. Diagnostic accuracy of ai for opportunistic screening of abdominal aortic aneurysm in CT: a systematic review and narrative synthesis. Diagnostics. 2022;12(12):3197. doi: 10.3390/diagnostics12123197 EDN: ERWYPX</mixed-citation></ref><ref id="B8"><label>8.</label><mixed-citation>Horiuchi D, Tatekawa H, Oura T, et al. ChatGPT’s diagnostic performance based on textual vs. visual information compared to radiologists’ diagnostic performance in musculoskeletal radiology. European Radiology. 2024;35(1):506–516. doi: 10.1007/s00330-024-10902-5 EDN: JAHWFM</mixed-citation></ref><ref id="B9"><label>9.</label><mixed-citation>Mitsuyama Y, Tatekawa H, Takita H, et al. Comparative analysis of GPT-4-based ChatGPT’s diagnostic performance with radiologists using real-world radiology reports of brain tumors. European Radiology. 2024;35(4):1938–1947. doi: 10.1007/s00330-024-11032-8 EDN: UHMLBQ</mixed-citation></ref><ref id="B10"><label>10.</label><mixed-citation>Kaya K, Gietzen C, Hahnfeldt R, et al. Generative Pre-trained Transformer 4 analysis of cardiovascular magnetic resonance reports in suspected myocarditis: A multicenter study. Journal of Cardiovascular Magnetic Resonance. 2024;26(2):101068. doi: 10.1016/j.jocmr.2024.101068 EDN: TSRLJX</mixed-citation></ref><ref id="B11"><label>11.</label><mixed-citation>Grolleau E, Couraud S, Jupin Delevaux E, et al. Incidental pulmonary nodules: Natural language processing analysis of radiology reports. Respiratory Medicine and Research. 2024;86:101136. doi: 10.1016/j.resmer.2024.101136 EDN: DHDPIX</mixed-citation></ref><ref id="B12"><label>12.</label><mixed-citation>Khoruzhaya AN, Kozlov DV, Arzamasov KM, Kremneva EI. Comparison of an ensemble of machine learning models and the BERT language model for analysis of text descriptions of brain CT reports to determine the presence of intracranial hemorrhage. Sovremennye tehnologii v medicine. 2024;16(1):27–36. doi: 10.17691/stm2024.16.1.03 EDN: AXXVVD</mixed-citation></ref><ref id="B13"><label>13.</label><mixed-citation>Han T, Adams LC, Bressem KK, et al. Comparative analysis of multimodal large language model performance on clinical vignette questions. JAMA. 2024;331(15):1320–1321. doi: 10.1001/jama.2023.27861 EDN: KPFLZG</mixed-citation></ref><ref id="B14"><label>14.</label><mixed-citation>Horiuchi D, Tatekawa H, Shimono T, et al. Accuracy of ChatGPT generated diagnosis from patient's medical history and imaging findings in neuroradiology cases. Neuroradiology. 2023;66(1):73–79. doi: 10.1007/s00234-023-03252-4 EDN: SRFGAA</mixed-citation></ref><ref id="B15"><label>15.</label><mixed-citation>Wataya T, Miura A, Sakisuka T, et al. Comparison of natural language processing algorithms in assessing the importance of head computed tomography reports written in Japanese. Japanese Journal of Radiology. 2024;42(7):697–708. doi: 10.1007/s11604-024-01549-9 EDN: VAKPBV</mixed-citation></ref><ref id="B16"><label>16.</label><mixed-citation>Cagnina A, Salihu A, Meier D, et al. Assessing the need for coronary angiography in high-risk non-ST-elevation acute coronary syndrome patients using artificial intelligence and computed tomography. The International Journal of Cardiovascular Imaging. 2024;41(1):55–61. doi: 10.1007/s10554-024-03283-9 EDN: JMBFSX</mixed-citation></ref><ref id="B17"><label>17.</label><mixed-citation>Gallifant J, Afshar M, Ameen S, et al. The TRIPOD-LLM reporting guideline for studies using large language models. Nature Medicine. 2025;31(1):60–69. doi: 10.1038/s41591-024-03425-5 EDN: KAPIXF</mixed-citation></ref><ref id="B18"><label>18.</label><mixed-citation>Tripathi S, Alkhulaifat D, Doo FX, et al. Development, evaluation, and assessment of large language models (DEAL) checklist: a technical report. NEJM AI. 2025;2(6). doi: 10.1056/AIp2401106</mixed-citation></ref><ref id="B19"><label>19.</label><mixed-citation>Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society Series B: Statistical Methodology. 1995;57(1):289–300. doi: 10.1111/j.2517-6161.1995.tb02031.x</mixed-citation></ref><ref id="B20"><label>20.</label><mixed-citation>Hollestein LM, Lo SN, Leonardi-Bee J, et al. MULTIPLE ways to correct for MULTIPLE comparisons in MULTIPLE types of studies. British Journal of Dermatology. 2021;185(6):1081–1083. doi: 10.1111/bjd.20600 EDN: QQWVVP</mixed-citation></ref><ref id="B21"><label>21.</label><mixed-citation>Collins GS, Moons KGM, Dhiman P, et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ. 2024;385:e078378. doi: 10.1136/bmj-2023-078378 EDN: WSTQKK</mixed-citation></ref><ref id="B22"><label>22.</label><mixed-citation>Cohen JF, Korevaar DA, Altman DG, et al. STARD 2015 guidelines for reporting diagnostic accuracy studies: explanation and elaboration. BMJ Open. 2016;6(11):e012799. doi: 10.1136/bmjopen-2016-012799</mixed-citation></ref><ref id="B23"><label>23.</label><mixed-citation>Bossuyt PM, Reitsma JB, Bruns DE, et al. STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies. BMJ. 2015;351:h5527. doi: 10.1136/bmj.h5527</mixed-citation></ref><ref id="B24"><label>24.</label><mixed-citation>Vasiliev YuA, Vlazimirsky AV, Omelyanskaya OV, et al. Methodology for testing and monitoring artificial intelligence-based software for medical diagnostics. Digital Diagnostics. 2023;4(3):252–267. doi: 10.17816/DD321971 EDN: UEDORU</mixed-citation></ref><ref id="B25"><label>25.</label><mixed-citation>Vasilev YuA, Bobrovskaya TM, Arzamasov KM, et al. Medical datasets for machine learning: fundamental principles of standartization and systematization. Manager Zdravookhranenia. 2023; (4):28–41. doi: 10.21045/1811-0185-2023-4-28-41 EDN: EPGAMD</mixed-citation></ref><ref id="B26"><label>26.</label><mixed-citation>Vinogradova IA, Nizovtsova LA, Omelyanskaya OV. Innovative strategic session in the scientific activity of the Center for Diagnostics and Telemedicine. Digital Diagnostics. 2022;3(4):414–420. doi: 10.17816/DD111833 EDN: DLRLVI</mixed-citation></ref><ref id="B27"><label>27.</label><mixed-citation>Kalinina ML, Svitachev AP, Biswas D, Vishnu P. Comparison of awareness and attitudes toward artificial intelligence among Russian- and English-speaking students at Orenburg State Medical University. Digital Diagnostics. 2023;4(1S):62–65. doi: 10.17816/DD430346 EDN: DIKOYA</mixed-citation></ref></ref-list></back></article>
