使用人工智能服务分析胸部X光片的局限
- 作者: Vasilev Y.A.1, Vladzymyrskyy A.V.1, Arzamasov K.M.1, Shulkin I.M.1, Astapenko E.V.1, Pestrenin L.D.1
-
隶属关系:
- Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies
- 期: 卷 5, 编号 3 (2024)
- 页面: 407-420
- 栏目: 原创性科研成果
- URL: https://journals.rcsi.science/DD/article/view/310027
- DOI: https://doi.org/10.17816/DD626310
- ID: 310027
如何引用文章
全文:
详细
论证。在放射诊断中,最早开始应用人工智能和积极使用至今的领域之一是胸部X光片。 然而,在使用人工智能(AI)技术解释这些研究时,放射科医生仍然每天面临着诸多局限,在做出医疗报告时必须考虑这些局限,这些局限必须受到开发人员的重视,以便进一步改进算法,提高效率。
目的。确定使用人工智能服务进行胸部X光片分析的局限,并评估这些局限的临床意义。
材料和方法。在分析155例患者胸部X光片时对人工智能服务结论与医疗报告不一致的病例进行回顾性分析。所有研究病例均来自莫斯科市统一医疗信息分析系统的统一放射信息服务。
结果。在被分析的155个差异病例中,48个(31.0%)为假阳性,78 个(50.3%)为假阴性。经专家审查发现其余29例(18.7%)为真阳性(27)或真阴性(2),因此这些病例被排除在进一步研究之外。在48个假阳性病例中,大多数(93.8%)是由于人工智能服务将胸部正常解剖结构(97.8%的病例)或导管阴影(2.2%的病例)误认为是气胸的体征。在假阴性研究中,临床显著性病理的漏诊比例为22.0%。这些病例中几乎一半(44.4%)与漏诊的肺结节有关。最常见的无临床意义的病理是肺钙化(60.9%)。
结论。在人工智能服务方面存在过度诊断的倾向。所有假阳性病例均与临床显著性病理的错误检测有关:气胸、肺结节和肺部阴影。在假阴性病例中,漏诊有临床意义显著性病理的比例很小,且不到四分之一。
作者简介
Yuriy A. Vasilev
Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies
Email: npcmr@zdrav.mos.ru
ORCID iD: 0000-0002-5283-5961
SPIN 代码: 4458-5608
MD, Cand. Sci. (Medicine)
俄罗斯联邦, MoscowAnton V. Vladzymyrskyy
Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies
Email: a.vladzimirsky@npcmr.ru
ORCID iD: 0000-0002-2990-7736
SPIN 代码: 3602-7120
MD, Dr. Sci. (Medicine), Professor
俄罗斯联邦, MoscowKirill M. Arzamasov
Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies
Email: ArzamasovKM@zdrav.mos.ru
ORCID iD: 0000-0001-7786-0349
SPIN 代码: 3160-8062
MD, Cand. Sci. (Medicine)
俄罗斯联邦, MoscowIgor M. Shulkin
Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies
Email: i.shulkin@npcmr.ru
ORCID iD: 0000-0002-7613-5273
SPIN 代码: 5266-0618
俄罗斯联邦, Moscow
Elena V. Astapenko
Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies
编辑信件的主要联系方式.
Email: AstapenkoEV1@zdrav.mos.ru
ORCID iD: 0009-0006-6284-2088
SPIN 代码: 7362-8553
俄罗斯联邦, Moscow
Lev D. Pestrenin
Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies
Email: PestreninLD@zdrav.mos.ru
ORCID iD: 0000-0002-1786-4329
SPIN 代码: 7193-7706
俄罗斯联邦, Moscow
参考
- Çallı E, Sogancioglu E, van Ginneken B, et al. Deep learning for chest X-ray analysis: A survey. Medical Image Analysis. 2021;72:102125. doi: 10.1016/j.media.2021.102125
- Vasilev YuA, Tyrov IA, Vladzymyrskyy AV, et al. A New Model of Organizing Mass Screening Based on Stand-Alone Artificial Intelligence Used for Fluorography Image Triage. Zdorov’e Naseleniya i Sreda Obitaniya. 2023;31(11):23–32. (In Russ.) doi: 10.35627/2219-5238/2023-31-11-23-32
- Akhter Y, Singh R, Vatsa M. AI-based radiodiagnosis using chest X-rays: A review. Frontiers in Big Data. 2023;6:1120989. doi: 10.3389/fdata.2023.1120989
- Fanni SC, Marcucci A, Volpi F, et al. Artificial Intelligence-Based Software with CE Mark for Chest X-ray Interpretation: Opportunities and Challenges. Diagnostics (Basel). 2023;13(12):2020. doi: 10.3390/diagnostics13122020
- Gusev AV, Vladzymyrskyy AV, Sharova DE, et al. Evolution of research and development in the field of artificial intelligence technologies for healthcare in the Russian Federation: results of 2021. Digital Diagnostics. 2022;3(3):178–194. doi: 10.17816/DD107367
- Kim J, Kim KH. Role of chest radiographs in early lung cancer detection. Translational Lung Cancer Research. 2020;9(3):522–531. doi: 10.21037/tlcr.2020.04.02
- Golubev NA, Ogryzko EV, Tyurina EM, et al. Features of the development of the radiology diagnostic service in the Russian Federation for 2014-2019. Current problems of health care and medical statistics. 2021;(2):356–376. doi: 10.24412/2312-2935-2021-2-356-376
- Wu JT, Wong KCL, Gur Y, et al. Comparison of Chest Radiograph Interpretations by Artificial Intelligence Algorithm vs Radiology Residents. JAMA Network Open. 2020;3(10):e2022779. doi: 10.1001/jamanetworkopen.2020.22779
- Miró Catalina Q, Fuster-Casanovas A, Solé-Casals J, Vidal-Alaball J. Developing an Artificial Intelligence Model for Reading Chest X-rays: Protocol for a Prospective Validation Study. JMIR Research Protocols. 2022;11(11):e39536. doi: 10.2196/39536
- Plesner LL, Müller FC, Nybing JD, et al. Autonomous Chest Radiograph Reporting Using AI: Estimation of Clinical Impact. Radiology. 2023;307(3):e222268. doi: 10.1148/radiol.222268
- Vasilev Yu, Vladzymyrskyy A, Omelyanskaya O, et al. AI-Based CXR First Reading: Current Limitations to Ensure Practical Value. Diagnostics (Basel). 2023;13(8):1430. doi: 10.3390/diagnostics13081430
- Driver CN, Bowles BS, Bartholmai BJ, Greenberg-Worisek AJ. Artificial Intelligence in Radiology: A Call for Thoughtful Application. Clinical and Translational Science. 2020;13(2):216–218. doi: 10.1111/cts.12704
- Yoo H, Kim EY, Kim H, et al. Artificial intelligence-based identification of normal chest radiographs: a simulation study in a multicenter health screening cohort. Korean Journal of Radiology. 2022;23(10):1009–1018. doi: 10.3348/kjr.2022.0189
- Suganuma N, Yoshida S, Takeuchi Y, et al. Artificial intelligence in quantitative chest imaging analysis for occupational lung disease. Seminars in Respiratory and Critical Care Medicine. 2023;44(3):362–369. doi: 10.1055/s-0043-1767760
- Brown C, Nazeer R, Gibbs A, et al. Breaking Bias: The role of artificial intelligence in improving clinical decision-making. Cureus. 2023;15(3):e36415. doi: 10.7759/cureus.36415
- Kaviani P, Kalra MK, Digumarthy SR, et al. Frequency of missed findings on chest radiographs (CXRs) in an international, multicenter study: application of AI to reduce missed findings. Diagnostics (Basel). 2022;12(10):2382. doi: 10.3390/diagnostics12102382
- de Groot PM, Carter BW, Abbott GF, Wu CC. Pitfalls in chest radiographic interpretation: blind spots. Seminars in Roentgenology. 2015;50(3):197–209. doi: 10.1053/j.ro.2015.01.008
- Gefter WB, Post BA, Hatabu H. Commonly missed findings on chest radiographs: causes and consequences. Chest. 2023;163(3):650–661. doi: 10.1016/j.chest.2022.10.039
- Vasilev YuA, Vladzimirsky AV, editors. Komp’yuternoe zrenie v luchevoj diagnostike: pervyj etap Moskovskogo eksperimenta. Moscow: Izdatelskie resheniya; 2022. (In Russ.)
- Morozov SP, Burenchev DV, Vladzimirsky AV, et al. Principy i pravila opisanij rezul’tatov luchevyh issledovanij. The series “Best practices of radiation and instrumental diagnostics”. Issue 97. Moscow: State Budget-Funded Health Care Institution of the City of Moscow “Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department”; 2021. (In Russ.)
- Choi YR, Yoon SH, Kim J, et al. Chest Radiography of Tuberculosis: Determination of Activity Using Deep Learning Algorithm. Tuberculosis and Respiratory Diseases. 2023;86(3):226–233. doi: 10.4046/trd.2023.0020
- Sun Z, Zhou J, Zhao L. Application status and problems summary of artificial intelligence in chest imaging. Asian Journal of Surgery. 2023;46(10):4437–4438. doi: 10.1016/j.asjsur.2023.04.100
- Bernstein MH, Atalay MK, Dibble EH, et al. Can incorrect artificial intelligence (AI) results impact radiologists, and if so, what can we do about it? A multi-reader pilot study of lung cancer detection with chest radiography. European Radiology. 2023;33(11):8263–8269. doi: 10.1007/s00330-023-09747-1
- Becker J, Decker JA, Römmele C, et al. Artificial Intelligence-based detection of pneumonia in chest radiographs. Diagnostics (Basel). 2022;12(6):1465. doi: 10.3390/diagnostics12061465
- Dasegowda G, Bizzo BC, Gupta RV, et al. Radiologist-trained AI model for identifying suboptimal chest-radiographs. Academic Radiology. 2023;30(12):2921–2930. doi: 10.1016/j.acra.2023.03.006
- Fanni SC, Greco G, Rossi S, et al. Role of artificial intelligence in oncologic emergencies: a narrative review. Exploration of Targeted Anti-tumor Therapy. 2023;4(2):344–354. doi: 10.37349/etat.2023.00138
- Hwang EJ, Goo JM, Nam JG, et al. Conventional versus artificial intelligence-assisted interpretation of chest radiographs in patients with acute respiratory symptoms in emergency department: a pragmatic randomized clinical trial. Korean Journal of Radiology. 2023;24(3):259–270. doi: 10.3348/kjr.2022.0651
- Tan H, Xu H, Yu N, et al. The value of deep learning-based computer aided diagnostic system in improving diagnostic performance of rib fractures in acute blunt trauma. BMC Medical Imaging. 2023;23(1):55. doi: 10.1186/s12880-023-01012-7
- Wu J, Liu N, Li X, et al. Convolutional neural network for detecting rib fractures on chest radiographs: a feasibility study. BMC Medical Imaging. 2023;23(1):18. doi: 10.1186/s12880-023-00975-x
- Lee HW, Jin KN, Oh S, et al. Artificial intelligence solution for chest radiographs in respiratory outpatient clinics: multicenter prospective randomized clinical trial. Annals of the American Thoracic Society. 2023;20(5):660–667. doi: 10.1513/AnnalsATS.202206-481OC
- Hillis JM, Bizzo BC, Mercaldo S, et al. Evaluation of an artificial intelligence model for detection of pneumothorax and tension pneumothorax in chest radiographs. JAMA Network Open. 2022;5(12):e2247172. doi: 10.1001/jamanetworkopen.2022.47172
补充文件
