The answer at our fingertips: Volume status in cirrhosis determined by machine learning and pulse oximeter waveform.
The answer at our fingertips: Volume status in cirrhosis determined by machine learning and pulse oximeter waveform.
复制标题
DOI：
10.14814/phy2.15223
复制
发表时间：
2022-03
影响因子：
2.5
通讯作者：
Levitsky J
中科院分区：
文献类型：
--
作者：
Mazumder NR;Kazen A;Carek A;Etemadi M;Levitsky J
我们研究的目的是确定来自简单脉搏血氧仪样设备的波形是否可用于准确评估肝硬化患者的血管内容量状态。肝硬化患者在心导管插入术当天进行了波形记录以及血清脑钠肽（BNP），其中测量了有创心脏压力。处理波形以生成用于机器学习模型的特征，以便预测充盈压力（回归）或将患者分类为容量过载或非容量过载（定义为LVEDP>15）。26例患者中有9例（35%）存在血管内容量超负荷。采用PPG特征的回归分析（R2 = 0.66）上级BNP（R2 = 0.22）。线性判别分析正确分类患者的准确性为78%，灵敏度为60%，阳性预测值为90%，AUROC为0.87。脉搏血氧波形的机器学习增强分析可以估计血管内容量超负荷，其准确性高于传统测量的BNP。管理肝硬化患者的一个主要挑战是，由于缺乏血管内容量状态的指标，难以进行治疗。在这项初步研究中，我们开发了一个模型，预测容量状态的变化，在屏气过程中的指尖光电容积脉搏波波形。
The objective of our study was to determine if the waveform from a simple pulse oximeter‐like device could be used to accurately assess intravascular volume status in cirrhosis. Patients with cirrhosis underwent waveform recording as well as serum brain natriuretic peptide (BNP) on the day of their cardiac catheterization where invasive cardiac pressures were measured. Waveforms were processed to generate features for machine learning models in order to predict the filling pressures (regression) or to classify the patients as volume overloaded or not (defined as an LVEDP>15). Nine of 26 patients (35%) had intravascular volume overload. Regression analysis using PPG features (R 2 = 0.66) was superior to BNP (R2 = 0.22). Linear discriminant analysis correctly classified patients with an accuracy of 78%, sensitivity of 60%, positive predictive value of 90%, and an AUROC of 0.87. Machine learning‐enhanced analysis of pulse ox waveforms can estimate intravascular volume overload with a higher accuracy than conventionally measured BNP. A major challenge in managing patients with cirrhosis is that it is difficult to dose treatments due to the lack of a metric for intravascular volume status. In this pilot study, we developed a model that predicted volume status using the changes to a fingertip photoplethysmography waveform during breath holding.