首页|New Findings from University of California Davis Update Understanding of Machine Learning (Improving the Precision of Shock Resuscitation By Predicting Fluid Re sponsiveness With Machine Learning and Arterial Blood Pressure Waveform Data)
New Findings from University of California Davis Update Understanding of Machine Learning (Improving the Precision of Shock Resuscitation By Predicting Fluid Re sponsiveness With Machine Learning and Arterial Blood Pressure Waveform Data)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Investigators discuss new findings in Machine Learning. According to news reporting out of Sacramento, California, by NewsRx editors, research stated, "Fluid bolus therapy (FBT) is fundamental to th e management of circulatory shock in critical care but balancing the benefits an d toxicities of FBT has proven challenging in individual patients. Improved pred ictors of the hemodynamic response to a fluid bolus, commonly referred to as a f luid challenge, are needed to limit non-beneficial fluid administration and to e nable automated clinical decision support and patient-specific precision critica l care management." Financial support for this research came from United States Department of Defens e. Our news journalists obtained a quote from the research from the University of C alifornia Davis, "In this study we retrospectively analyzed data from 394 fluid boluses from 58 pigs subjected to either hemorrhagic or distributive shock. All animals had continuous blood pressure and cardiac output monitored throughout th e study. Using this data, we developed a machine learning (ML) model to predict the hemodynamic response to a fluid challenge using only arterial blood pressure waveform data as the input. A Random Forest binary classifier referred to as th e ML fluid responsiveness algorithm (MLFRA) was trained to detect fluid responsi veness (FR), defined as a>= 15% change in cardiac stroke volume after a fluid challenge. We then compared its performance to pulse pressure variation, a commonly used metric of FR. Model performance was assessed using the area under the receiver operating characteristic curve (AURO C), confusion matrix metrics, and calibration curves plotting predicted probabil ities against observed outcomes. Across multiple train/test splits and feature s election methods designed to assess performance in the setting of small sample s ize conditions typical of large animal experiments, the MLFRA achieved an averag e AUROC, recall (sensitivity), specificity, and precision of 0.82, 0.86, 0.62. a nd 0.76, respectively. In the same datasets, pulse pressure variation had an AUR OC, recall, specificity, and precision of 0.73, 0.91, 0.49, and 0.71, respective ly. The MLFRA was generally well-calibrated across its range of predicted probab ilities and appeared to perform equally well across physiologic conditions. Thes e results suggest that ML, using only inputs from arterial blood pressure monito ring, may substantially improve the accuracy of predicting FR compared to the us e of pulse pressure variation."
SacramentoCaliforniaUnited StatesN orth and Central AmericaCyborgsEmerging TechnologiesMachine LearningUniv ersity of California Davis
NETL
NSTL
