新型超声乳腺肿块计算机辅助分类系统的可行性与诊断价值:单中心初步研究
本研究纳入246个乳腺肿块,由放射科医师与计算机辅助分类(CAC)系统按BI-RADS词典评估超声特征及最终分类。CAC系统融合多视角与彩色多普勒图像,分无临床变量(SmartBreast)和结合临床变量(m-CAC)两版。结果:CAC与医师在肿块形态、方位、边缘、后方特征、钙化及血供方面一致性显著,以BI-RADS 4a为截断值,m-CAC较SmartBreast提高特异度(52.5% vs 25%)和准确率(73.98% vs 62.6%)且敏感度无显著下降;m-CAC与医师特异度和准确率相近,但敏感度较低(94.44% vs 100%)。
要点
- CAC系统与医师在BI-RADS征象判读上呈中高度一致(κ 0.379–0.745)。
- 结合临床变量的m-CAC可提升特异度至52.5%,敏感度94.44%。
- CAC系统有潜力模拟放射科医师对乳腺病灶的分类决策行为。
关键影像表现/征象
- CAC系统与医师在肿块形态(κ=0.673)、方位(κ=0.682)、边缘(κ=0.622)上一致性显著。
- CAC系统与医师在肿块内钙化(κ=0.709)和血供(κ=0.745)上一致性显著。
- CAC系统与医师在后方特征(κ=0.629)上一致性显著,回声模式一致性一般(κ=0.379)。
- CAC系统与医师在BI-RADS最终评估分类上一致性中等(κ=0.575)。
诊断标准与关键数值
- 以BI-RADS 4a为截断值,m-CAC系统特异度52.5%,SmartBreast特异度25%(p<0.0001)。
- 以BI-RADS 4a为截断值,m-CAC系统准确率73.98%,SmartBreast准确率62.6%(p=0.0002)。
- 以BI-RADS 4a为截断值,m-CAC系统敏感度94.44%,SmartBreast敏感度98.41%(p=0.1250)。
- 以BI-RADS 4a为截断值,m-CAC系统与医师特异度(52.5% vs 45.83%,p=0.2430)和准确率(73.98% vs 73.58%,p=1.0000)相近,但敏感度较低(94.44% vs 100%,p=0.0156)。
临床应用与阅片要点
- CAC系统可作为BI-RADS 4a截断值下辅助分类工具,结合临床变量(m-CAC)可提高特异度、减少不必要活检。
- 阅片时注意CAC对回声模式判读一致性一般(κ=0.379),该征象仍需医师重点复核。
- m-CAC敏感度低于医师,阴性或低分类病灶不应仅凭CAC结果排除恶性。
局限性
- 单中心初步测试研究,样本量有限(246个肿块),结论需多中心验证。
- 仅与一名放射科医师的判读进行比较,缺乏观察者间一致性评估。
- 为可行性初步研究,未报告病理金标准及前瞻性诊断效能验证。
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
AIM: To introduce a novel computer-aided classification (CAC) system and investigate the feasibility of characterising and diagnosing breast masses on ultrasound (US).
MATERIALS AND METHODS: A total of 246 breast masses were included. US features and the final assessment categories of the breast masses were analysed by a radiologist and the CAC system according to the Breast Imaging Reporting and Data System (BI-RADS) lexicon. The CAC system evaluated the BI-RADS assessment from the fusion of multi-view and colour Doppler US images without (SmartBreast) or with combining clinical variables (m-CAC system). The diagnostic performance and agreement of US characteristics between the radiologist and the CAC system were compared.
RESULTS: The agreement between the radiologist and the CAC system was substantial for mass shape (κ = 0.673), orientation (κ = 0.682), margin (κ = 0.622), posterior features (κ = 0.629), calcifications in a mass (κ = 0.709) and vascularity (κ = 0.745), fair for echo pattern (κ = 0.379), and moderate for BI-RADS assessment (κ = 0.575). With BI-RADS 4a as the cut-off value, the specificity (52.5% versus 25%, p<0.0001) and accuracy (73.98% versus 62.6%, p=0.0002) of the m-CAC system were improved without significant loss of sensitivity (94.44% versus 98.41%, p=0.1250) compared with the SmartBreast. The m-CAC system showed similar specificity (52.5% versus 45.83%, p=0.2430) and accuracy (73.98% versus 73.58%, p=1.0000) as the radiologist, but a lower sensitivity (94.44% versus 100%, p=0.0156).
CONCLUSION: The CAC system showed an acceptable agreement with the radiologist for characterisation of breast lesions. It has the potential to mimic the decision-making behaviour of radiologists for the classification of breast lesions.
引用
- Gu, Xu, Liu, et al. The feasibility of a novel computer-aided classification system for the characterisation and diagnosis of breast masses on ultrasound: a single-centre preliminary test study. Clin Radiol. 2023. PMID: 37069025. doi:10.1016/j.crad.2023.03.011
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。