用于乳腺癌检测的人工智能系统在查乳房扫描器挪威的乳房扫描仪的乳房扫描仪上的表现
Marthe Larsen1, Camilla F Olstad1, Christoph I Lee1
1From the Section for Breast Cancer Screening (M.L., C.F.O., S.H.) and Department of Register Informatics (S.A., J.F.N.), Cancer Registry of Norway, Norwegian Institute of Public Health, PO 5313, Majorstuen, 0304 Oslo, Norway; Department of Radiology, University of Washington School of Medicine, Seattle, Wash (C.I.L.); Department of Health Systems and Population Health, University of Washington School of Public Health, Seattle, Wash (C.I.L.); Department of Radiology, Vestre Viken Hospital Trust, Drammen, Norway (T.H.); Department of Radiology, Ålesund Hospital, Møre og Romsdal Hospital Trust, Ålesund, Norway (S.R.H.); Department of Circulation, Medical Imaging, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology, Trondheim, Norway (S.R.H.); Department of Radiology, Østfold Hospital Trust, Kalnes, Norway (M.A.M.); Institute of Clinical Medicine, University of Oslo, Oslo, Norway (M.A.M.); SPKI-The Norwegian Centre for Clinical Artificial Intelligence, University Hospital of North Norway, Tromsø, Norway (K.Ø.M.); Department of Clinical Medicine, Faculty of Health Sciences (K.Ø.M.), Department of Physics and Technology, Faculty of Science and Technology (J.F.N.), and Department of Health and Care Sciences, Faculty of Health Sciences (S.H.), UiT-The Arctic University of Norway, Tromsø, Norway; Department of Radiology and Nuclear Medicine, St Olavs University Hospital, Trondheim, Norway (H.L.H.); Department of Radiology, Hospital of Southern Norway, Kristiansand, Norway (H.S.S.); Department of Radiology, Innlandet Hospital Trust, Hamar, Norway (M.S.); and Department of Radiology, Innlandet Hospital Trust, Lillehammer, Norway (Å.Ø.S.).
一个商业人工智能 (AI) 系统在通过乳房扫描检测乳腺癌方面表现出了很高的性能. 这种人工智能工具显示出对乳房影像分类的潜力,旨在减少放射科医生的工作量.
科学领域:
- 放射学和医学成像学 医学成像学
- 医疗保健中的人工智能
- 在瘤学瘤学.
背景情况:
- 乳房显微镜是乳腺癌查的一个关键工具.
- 放射科医生的工作量可能会影响查效率和准确性.
- 人工智能系统正在开发,以帮助医疗图像分析.
研究的目的:
- 评估商业AI系统的独立乳腺癌检测性能.
- 评估AI在各种风险分数门上的表现.
- 探索人工智能系统在乳房影像分类方面的潜力.
主要方法:
- 从挪威的BreastScreen挪威 (2004-2018) 进行了661695次数字乳房扫描检查的回顾性分析.
- 包括 3,807 种查检测的癌症和 1,110 种间隔性乳腺癌.
- 利用人工智能系统的持续风险评分来计算曲线下的面积 (AUC) 和癌症检测率.
主要成果:
- 人工智能系统在查和间隔检测的癌症中实现了0.93的AUC,在查检测的癌症中达到0.97.
- 在10%的阳性值下,AI识别了92.0%的查检测和44.6%的间隔癌症,将68.5%的假阳性归类为负.
- 在50%的阳性值下,AI识别了99.3%的屏幕检测和85.2%的间隔癌症,将17.0%的虚假阳性归类为负.
结论:
- 人工智能系统在检测乳腺癌时表现出很高的性能,检测乳腺癌是在进行乳腺查后两年内进行的.
- 人工智能系统在乳房影像分类方面具有潜在的实用性,特别是识别低风险病例.
- 这种人工智能应用程序可以帮助减轻乳腺癌查计划中的放射科医生工作量.


