将人工智能柠檬水从数据柠檬中挤出:利用公共心尖超声数据库创建剑突下视觉估测自动射血分数机器学习算法
研究利用仅含心尖四腔心图像的斯坦福EchoNet-Dynamic数据库(10,036个视频)训练机器学习算法,通过旋转和翻转图像模拟剑突下切面,并在Harbor-UCLA的615个剑突下视频上测试。算法平均绝对误差在仅用原始图像时为23.0,加入90°顺时针旋转后为16.7,再加入水平翻转后为16.6;Bland-Altman分析显示在EF 40-45%时一致性最弱。作者认为数据操作可模拟不同超声切面,可能有助于克服床旁超声数据库匮乏的问题。
要点
- 仅用公共心尖四腔心数据训练剑突下切面EF估测算法,平均绝对误差最低为16.6。
- 图像旋转和水平翻转可部分模拟剑突下切面,改善算法在剑突下视频上的表现。
- Bland-Altman分析显示算法在EF 40-45%范围一致性最弱。
核心结论
- 研究者成功利用无关超声切面数据,通过数据操作模拟不同超声检查,训练出平均绝对误差尚可的床旁超声机器学习算法;这可能对未来床旁超声算法设计以帮助克服数据库匮乏具有意义。
研究设计与样本
- 研究采用长短期记忆算法进行图像分析;使用斯坦福EchoNet-Dynamic数据库中10,036个心尖四腔心视频(含计算的确切EF)测试3种机器学习训练方案;并在Harbor-UCLA获取的615个剑突下视频(EF以5%范围计算)上做真实世界测试;对Harbor-UCLA的EF范围进行1000次随机化点估计以补偿机器学习与Harbor-UCLA EF不匹配,获得平均绝对误差比较均值并行Bland-Altman分析。
诊断标准与关键数值
- 仅用原始心尖四腔心视频训练时,机器学习算法EF平均绝对误差估计为23.0,范围22.8-23.3。
- 用原始加90°顺时针旋转视频训练时,平均绝对误差为16.7,范围16.5-16.9。
- 用原始、90°顺时针旋转加水平翻转视频训练时,平均绝对误差为16.6,范围16.3-16.8。
- Bland-Altman分析显示在EF 40-45%时一致性最弱。
临床应用与阅片要点
- 作者认为数据操作模拟不同超声切面可能对未来床旁超声算法设计以帮助克服床旁超声数据库匮乏具有意义。
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
OBJECTIVES: A paucity of point-of-care ultrasound (POCUS) databases limits machine learning (ML). Assess feasibility of training ML algorithms to visually estimate left ventricular ejection fraction (EF) from a subxiphoid (SX) window using only apical 4-chamber (A4C) images.
METHODS: Researchers used a long-short-term-memory algorithm for image analysis. Using the Stanford EchoNet-Dynamic database of 10,036 A4C videos with calculated exact EF, researchers tested 3 ML training permeations. First, training on unaltered Stanford A4C videos, then unaltered and 90° clockwise (CW) rotated videos and finally unaltered, 90° rotated and horizontally flipped videos. As a real-world test, we obtained 615 SX videos from Harbor-UCLA (HUCLA) with EF calculations in 5% ranges. Researchers performed 1000 randomizations of EF point estimation within HUCLA EF ranges to compensate for ML and HUCLA EF mismatch, obtaining a mean value for absolute error (MAE) comparison and performed Bland-Altman analyses.
RESULTS: The ML algorithm EF mean MAE was estimated at 23.0, with a range of 22.8-23.3 using unaltered A4C video, mean MAE was 16.7, with a range of 16.5-16.9 using unaltered and 90° CW rotated video, mean MAE was 16.6, with a range of 16.3-16.8 using unaltered, 90° CW rotated and horizontally flipped video training. Bland-Altman showed weakest agreement at 40-45% EF.
CONCLUSIONS: Researchers successfully adapted unrelated ultrasound window data to train a POCUS ML algorithm with fair MAE using data manipulation to simulate a different ultrasound examination. This may be important for future POCUS algorithm design to help overcome a paucity of POCUS databases.
引用
- Blaivas, Blaivas, Campbell, et al. Making Artificial Intelligence Lemonade Out of Data Lemons: Adaptation of a Public Apical Echo Database for Creation of a Subxiphoid Visual Estimation Automatic Ejection Fraction Machine Learning Algorithm. J Ultrasound Med. 2021. PMID: 34820867. doi:10.1002/jum.15889
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。