基于人工智能的CT扫描检测急腹症患者气腹:一项临床诊断准确性研究
这项回顾性诊断准确性研究纳入2019年1月1日至9月25日丹麦Herlev和Gentofte医院急高危腹部患者队列331例,以胃肠放射科顾问医师评级为参考标准,评估AI算法检测气腹的效能;气腹31例(9%),AUC为0.77(95%CI 0.66-0.87),在99%特异度下敏感度52%(16/31,95%CI 29-65%),排除少量游离气体(<0.25 mL)后AUC升至0.96(95%CI 0.89-1.0),99%特异度下敏感度81%(13/16),结论为AI算法敏感度低但特异度极高,支持其用于排除气腹。
要点
- AI算法检测气腹的AUC为0.77(95%CI 0.66-0.87),敏感度低而特异度极高。
- 在99%特异度下,敏感度52%(16/31),阳性似然比52(95%CI 16-165)。
- 排除<0.25 mL少量游离气体后AUC升至0.96(95%CI 0.89-1.0),99%特异度下敏感度81%(13/16)。
核心结论
- AI算法在临床环境中检测CT气腹敏感度低但特异度极高,支持其用于排除气腹的作用。
研究设计与样本
- 回顾性诊断准确性研究,纳入丹麦Herlev和Gentofte医院2019年1月1日至2019年9月25日急高危腹部患者连续队列331例,参考标准为胃肠放射科顾问医师对气腹(无、少量、中量、大量)的评级,指数试验为基于滑动窗口方法和深度循环神经网络核心的新型AI算法,主要结局为ROC曲线下面积(AUC)。
诊断标准与关键数值
- 气腹31例(9%),其中大量16例、中量7例、少量8例;AUC 0.77(95%CI 0.66-0.87);在特异度99%(297/300,95%CI 97-100%)下,敏感度52%(16/31,95%CI 29-65%),阳性似然比52(95%CI 16-165);排除<0.25 mL少量游离气体后AUC升至0.96(95%CI 0.89-1.0),99%特异度下敏感度81%(13/16),阳性似然比82(95%CI 27-254)。
临床应用与阅片要点
- AI算法敏感度低但特异度极高,支持其用于排除气腹;排除<0.25 mL少量游离气体后AUC升至0.96(95%CI 0.89-1.0),99%特异度下敏感度81%(13/16)。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
PURPOSE: The primary aim was to investigate the diagnostic performance of an Artificial Intelligence (AI) algorithm for pneumoperitoneum detection in patients with acute abdominal pain who underwent an abdominal CT scan.
METHOD: This retrospective diagnostic test accuracy study used a consecutive patient cohort from the Acute High-risk Abdominal patient population at Herlev and Gentofte Hospital, Denmark between January 1, 2019 and September 25, 2019. As reference standard, all studies were rated for pneumoperitoneum (subgroups: none, small, medium, and large amounts) by a gastrointestinal radiology consultant. The index test was a novel AI algorithm based on a sliding window approach with a deep recurrent neural network at its core. The primary outcome was the area under the curve (AUC) of the receiver operating characteristic (ROC).
RESULTS: Of 331 included patients (median age 68 years (Range 19-100; 180 women)) 31 patients (9%) had pneumoperitoneum (large: 16, moderate: 7, small: 8). The AUC was 0.77 (95% CI 0.66-0.87). At a specificity of 99% (297/300, 95% CI: 97-100%), sensitivity was 52% (16/31, 95% CI 29-65%), and positive likelihood ratio was 52 (95% CI 16-165). When excluding cases with smaller amounts of free air (<0.25 mL) the AUC increased to 0.96 (95% CI 0.89-1.0). At 99% specificity, sensitivity was 81% (13/16) and positive likelihood ratio was 82 (95% CI 27 - 254).
CONCLUSIONS: An AI algorithm identified pneumoperitoneum on CT scans in a clinical setting with low sensitivity but very high specificity, supporting its role for ruling in pneumoperitoneum.
引用
- Brejnebøl, Nielsen, Taubmann, et al. Artificial Intelligence based detection of pneumoperitoneum on CT scans in patients presenting with acute abdominal pain: A clinical diagnostic test accuracy study. Eur J Radiol. 2022. PMID: 35259709. doi:10.1016/j.ejrad.2022.110216
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。