Background Radiology practices have a high volume of unremarkable chest radiographs and artificial intelligence (AI) could possibly improve workflow by providing an automatic report. Purpose To estimate the proportion of unremarkable chest radiographs, where AI can correctly exclude pathology (ie, specificity) without increasing diagnostic errors. Materials and Methods In this retrospective study, consecutive chest radiographs in unique adult patients (≥18 years of age) were obtained January 1-12, 2020, at four Danish hospitals. Exclusion criteria included insufficient radiology reports or AI output error. Two thoracic radiologists, who were blinded to AI output, labeled chest radiographs as "remarkable" or "unremarkable" based on predefined unremarkable findings (reference standard). Radiology reports were classified similarly. A commercial AI tool was adapted to output a chest radiograph "remarkableness" probability, which was used to calculate specificity at different AI sensitivities. Chest radiographs with missed findings by AI and/or the radiology report were graded by one thoracic radiologist as critical, clinically significant, or clinically insignificant. Paired proportions were compared using the McNemar test. Results A total of 1961 patients were included (median age, 72 years [IQR, 58-81 years]; 993 female), with one chest radiograph per patient. The reference standard labeled 1231 of 1961 chest radiographs (62.8%) as remarkable and 730 of 1961 (37.2%) as unremarkable. At 99.9%, 99.0%, and 98.0% sensitivity, the AI had a specificity of 24.5% (179 of 730 radiographs [95% CI: 21, 28]), 47.1% (344 of 730 radiographs [95% CI: 43, 51]), and 52.7% (385 of 730 radiographs [95% CI: 49, 56]), respectively. With the AI fixed to have a similar sensitivity as radiology reports (87.2%), the missed findings of AI and reports had 2.2% (27 of 1231 radiographs) and 1.1% (14 of 1231 radiographs) classified as critical (P = .01), 4.1% (51 of 1231 radiographs) and 3.6% (44 of 1231 radiographs) classified as clinically significant (P = .46), and 6.5% (80 of 1231) and 8.1% (100 of 1231) classified as clinically insignificant (P = .11), respectively. At sensitivities greater than or equal to 95.4%, the AI tool exhibited less than or equal to 1.1% critical misses. Conclusion A commercial AI tool used off-label could correctly exclude pathology in 24.5%-52.7% of all unremarkable chest radiographs at greater than or equal to 98% sensitivity. The AI had equal or lower rates of critical misses than radiology reports at sensitivities greater than or equal to 95.4%. These results should be confirmed in a prospective study. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Yoon and Hwang in this issue.

中文翻译：

---

使用人工智能识别不明显的胸部X光照片以进行自动报告。


背景放射学实践有大量不明显的胸部 X 光片，人工智能 (AI) 可以通过提供自动报告来改进工作流程。目的 估计不明显的胸部 X 光片的比例，其中 AI 可以正确排除病理（即特异性）而不增加诊断错误。材料和方法在这项回顾性研究中，2020 年 1 月 1 日至 12 日在丹麦四家医院获得了特定成年患者（≥18 岁）的连续胸片。排除标准包括放射学报告不足或人工智能输出错误。两名对人工智能输出不知情的胸部放射科医生根据预先定义的不显着结果（参考标准）将胸部 X 光片标记为“显着”或“不显着”。放射学报告的分类类似。商业人工智能工具被用来输出胸片“显着性”概率，用于计算不同人工智能灵敏度下的特异性。一位胸部放射科医生将人工智能和/或放射学报告漏检的胸部 X 光照片分级为严重、有临床意义或无临床意义。使用麦克尼马尔检验比较配对比例。结果 总共纳入 1961 名患者（中位年龄 72 岁 [IQR，58-81 岁]；993 名女性），每名患者拍摄一张胸片。参考标准将 1961 年胸片中的 1231 幅 (62.8%) 标记为显着，将 1961 年中的 730 幅 (37.2%) 标记为不显着。在灵敏度为 99.9%、99.0% 和 98.0% 时，AI 的特异性为 24.5%（730 张放射照片中的 179 张 [95% CI: 21, 28]）、47.1%（730 张放射照片中的 344 张 [95% CI: 43, 51]）和 52.7%（730 张 X 光照片中的 385 张 [95% CI：49, 56]）。 由于 AI 固定为与放射学报告具有相似的敏感性 (87.2%)，因此 AI 和报告的遗漏发现有 2.2%（1231 张射线照片中的 27 幅）和 1.1%（1231 张射线照片中的 14 幅）被分类为严重（P = .01） ）、4.1%（1231 张 X 光照片中的 51 张）和 3.6%（1231 张 X 光照片中的 44 张）被分类为有临床意义（P = .46），6.5%（1231 张 X 光照片中的 80 张）和 8.1%（1231 张 X 光照片中的 100 张）被分类为临床不显着(P = .11) 分别。当灵敏度大于或等于 95.4% 时，AI 工具的关键失误率小于或等于 1.1%。结论 在标签外使用的商业 AI 工具可以正确排除所有不明显胸部 X 线照片中 24.5%-52.7% 的病理，灵敏度大于或等于 98%。在灵敏度大于或等于 95.4% 的情况下，AI 的严重失误率与放射学报告相同或更低。这些结果应该在前瞻性研究中得到证实。 © RSNA，2024 本文提供补充材料。另请参阅本期 Yoon 和 Hwang 的社论。