人工智能用于胸部X线分诊以减少放射科医生工作负荷
该回顾性研究纳入2017年6月1日至30日门诊初次胸部X线摄影患者1964例,采用随机交叉方式比较有无深度学习检测算法(DLD)辅助阅片,并模拟50%工作列表缩减。结果显示50%缩减场景下敏感度非劣于全部阅片(差异0.8%,95%CI -3.8%至5.5%,p=0.01),特异度从86.7%升至90.4%(p<0.001)。结论认为基于DLD的分诊可能大幅减少工作负荷而不降低敏感度,同时提高特异度。
要点
- 50%工作列表缩减模拟下敏感度非劣(差异0.8%,95%CI -3.8%至5.5%),特异度从86.7%升至90.4%。
- 探索性分析中工作负荷缩减至70%时敏感度仍非劣,差异为0%。
- AI单独判读错误诊断的法律责任问题在临床实施前仍需明确。
核心结论
- 基于深度学习检测算法的分诊可能大幅减少工作负荷而不降低敏感度,同时提高特异度。
研究设计与样本
- 回顾性研究纳入2017年6月1日至30日门诊初次胸部X线摄影患者1964例,阅片顺序按随机交叉方式确定,比较有无商业DLD(检测9种征象)辅助阅片,并模拟50%工作列表缩减,以5%为非劣效界值。
诊断标准与关键数值
- DLD检测9种放射学征象:肺不张、钙化、心脏增大、实变、纤维化、结节、气胸、胸腔积液、气腹;50%工作列表缩减场景下按概率评分降序排序,后半视为阴性检查。
临床应用与阅片要点
- 50%工作列表缩减模拟中敏感度非劣(差异0.8%,95%CI -3.8%至5.5%,p=0.01),特异度从86.7%升至90.4%(p<0.001);探索性分析中缩减至70%时敏感度仍非劣(差异0%)。
局限性
- 原文为回顾性研究。
- 基于AI单独判读的错误诊断的法律责任问题在临床实施前仍需明确。
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
OBJECTIVES: To evaluate whether deep learning-based detection algorithms (DLD)-based triaging can reduce outpatient chest radiograph interpretation workload while maintaining noninferior sensitivity.
METHODS: This retrospective study included patients who underwent initial chest radiography at the outpatient clinic between June 1 and June 30, 2017. Readers interpreted radiographs with/without a commercially available DLD that detects nine radiologic findings (atelectasis, calcification, cardiomegaly, consolidation, fibrosis, nodules, pneumothorax, pleural effusion, and pneumoperitoneum). The reading order was determined in a randomized, crossover manner. The radiographs were classified into negative and positive examinations. In a 50% worklist reduction scenario, radiographs were sorted in descending order of probability scores: the lower half was regarded as negative exams, while the remaining were read with DLD by radiologists. The primary analysis evaluated noninferiority in sensitivity between radiologists reading all radiographs and simulating a 50% worklist reduction, with the inferiority margin of 5%. The specificities were compared using McNemar's test.
RESULTS: The study included 1964 patients (median age [interquartile range], 55 years [40-67 years]). The sensitivity was 82.6% (195 of 236; 95% CI: 77.5%, 87.3%) when readers interpreted all chest radiographs without DLD and 83.5% (197 of 236; 95% CI: 78.8%, 88.1%) in the 50% worklist reduction scenario. The difference in sensitivity was 0.8% (95% CI: - 3.8%, 5.5%), establishing noninferiority of 50% worklist reduction (p = 0.01). The specificity increased from 86.7% (1498 of 1728) to 90.4% (1562 of 1728) (p < 0.001) with DLD-based triage.
CONCLUSION: Deep learning-based triaging may substantially reduce workload without lowering sensitivity while improving specificity.
CLINICAL RELEVANCE STATEMENT: Substantial workload reduction without lowering sensitivity was feasible using deep learning-based triaging of outpatient chest radiograph; however, the legal responsibility for incorrect diagnoses based on AI-standalone interpretation remains an issue that should be defined before clinical implementation.
KEY POINTS: • A 50% workload reduction simulation using deep learning-based detection algorithm maintained noninferior sensitivity while improving specificity. • The CT recommendation rate significantly decreased in the disease-negative patients, whereas it slightly increased in the disease-positive group without statistical significance. • In the exploratory analysis, the noninferiority of sensitivity was maintained until 70% of the workload was reduced; the difference in sensitivity was 0%.
引用
- Yoon SH, Park S, Jang S, et al. Use of artificial intelligence in triaging of chest radiographs to reduce radiologists' workload. Eur Radiol. 2023. PMID: 37615766. doi:10.1007/s00330-023-10124-1
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。