联合使用两种基于人工智能的算法进行乳腺X线摄影分诊:一项回顾性模拟研究
本研究分析3012例筛查或诊断性乳腺X线摄影(含213例癌症),用两种商用AI算法将病例分为高风险(前10%)、低风险(后20%)或不确定,并模拟五种联合分诊场景,比较敏感度、特异度、需复核片数和异常判读率。联合AI在敏感或特异模式上优于单一AI或人工判读,同时减少需放射科医生复核的片数。
要点
- 联合两种AI算法的分诊场景可提高敏感度或特异度,并减少需放射科医生复核的乳腺X线片数。
- 敏感模式敏感度84%,需复核片数减少18.3%,异常判读率23.3%。
- 特异模式特异度87.7%,需复核片数减少41.7%,异常判读率17%。
核心结论
- 在该三级中心癌症富集数据集中,联合两种AI算法可提高乳腺X线判读的敏感度或特异度,同时减少需放射科医生复核的片数。
- 场景选择应考虑临床需求,并需要在筛查人群中验证。
研究设计与样本
- 回顾性模拟研究,共分析3012例筛查或诊断性乳腺X线摄影,其中213例癌症。
- 使用两种商用AI算法(AI-1、AI-2),按恶性可能性将病例分为高风险(前10%)、低风险(后20%)或不确定。
- 设定五种联合AI分诊场景(敏感模式、特异模式、保守模式、序贯模式A和B),决定病例是自动召回、判为阴性还是转诊放射科医生判读;采用McNemar检验与单一AI及人工判读比较。
诊断标准与关键数值
- 敏感模式敏感度84%,优于单一AI(p=0.03 [AI-1]、0.01 [AI-2])和人工判读(p=0.03),需复核片数减少18.3%(异常判读率23.3%)。
- 特异模式特异度87.7%,优于单一AI(p<0.001 [AI-1、AI-2]),与人工判读相当(p=0.37),需复核片数减少41.7%(异常判读率17%)。
- 保守模式、序贯模式A和B的敏感度分别为82.2%、80.8%、80.3%,与单一AI或人工判读相当(p>0.05,所有比较),需复核片数分别减少9.8%、49.8%、49.8%(异常判读率分别为18.6%、21.6%、21.7%)。
局限性
- 本研究为来自三级中心的癌症富集数据集。
- 场景选择应在筛查人群中验证。
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 triaging scenarios involving two commercial AI algorithms to enhance mammography interpretation and reduce workload.
MATERIALS AND METHODS: A total of 3012 screening or diagnostic mammograms, including 213 cancer cases, were analyzed using two AI algorithms (AI-1, AI-2) and categorized as "high-risk" (top 10%), "minimal-risk" (bottom 20%), or "indeterminate" based on malignancy likelihood. Five triaging scenarios of combined AI use (Sensitive, Specific, Conservative, Sequential Modes A and B) determined whether cases would be autonomously recalled, classified as negative, or referred for radiologist interpretation. Sensitivity, specificity, number of mammograms requiring review, and abnormal interpretation rate (AIR) were compared against single AIs and manual reading using McNemar's test.
RESULTS: Sensitive Mode achieved 84% sensitivity, outperforming single AI (p = 0.03 [AI-1], 0.01 [AI-2]) and manual reading (p = 0.03), with an 18.3% reduction in mammograms requiring review (AIR, 23.3%). Specific Mode achieved 87.7% specificity, exceeding single AI (p < 0.001 [AI-1, AI-2]) and comparable to manual reading (p = 0.37), with a 41.7% reduction in mammograms requiring review (AIR, 17%). Conservative and Sequential Modes A and B achieved sensitivities of 82.2%, 80.8%, and 80.3%, respectively, comparable to single AI or manual reading (p > 0.05, all), with reductions of 9.8%, 49.8%, and 49.8% in mammograms requiring review (AIRs, 18.6%, 21.6%, 21.7%).
CONCLUSION: Combining two AI algorithms improved sensitivity or specificity in mammography interpretation while reducing mammograms requiring radiologist review in this cancer-enriched dataset from a tertiary center. Scenario selection should consider clinical needs and requires validation in a screening population.
KEY POINTS: Question AI algorithms have the potential to improve workflow efficiency by triaging mammograms. Combining algorithms trained under different conditions may offer synergistic benefits. Findings The combined use of two commercial AI algorithms for triaging mammograms improved sensitivity or specificity, depending on the scenario, while also reducing mammograms requiring radiologist review. Clinical relevance Integrating two commercial AI algorithms could enhance mammography interpretation over using a single AI for triaging or manual reading.
引用
- Kim HJ, Kim HH, Eom HJ, et al. Combined use of two artificial intelligence-based algorithms for mammography triaging: a retrospective simulation study. Eur Radiol. 2025. PMID: 40839134. doi:10.1007/s00330-025-11849-x
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。