Using Transfer Learning for Improved Mortality Prediction in a Data-Scarce Hospital Setting.
Using Transfer Learning for Improved Mortality Prediction in a Data-Scarce Hospital Setting.
复制标题
DOI：
10.1177/1178222617712994
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
Das R
中科院分区：
文献类型：
--
作者：
Desautels T;Calvert J;Hoffman J;Mao Q;Jay M;Fletcher G;Barton C;Chettipally U;Kerem Y;Das R
基于算法的临床决策支持（CDS）系统将患者来源的健康数据与感兴趣的结果（例如住院死亡率）关联起来。然而，这种关联的质量往往取决于特定地点的培训数据的可用性。如果没有足够的数据量，底层的统计装置就无法区分有用的模式和噪声，结果可能表现不佳。这种初始训练数据负担限制了基于机器学习的风险评分系统的广泛使用。在这项研究中，我们实现了一种统计迁移学习技术，它使用一个大的“源”数据集，以大大减少在训练数据稀缺的“目标”站点上表现良好所需的数据量。我们使用AutoTriage（一种死亡率预测算法）对Beth Israel Deaconess Medical Center（来源）的患者图表和加州大学弗朗西斯科医学中心（目标机构）的48249名成人住院患者进行了测试。我们发现，在目标集上超过0.80的受试者工作特征（AUROC）面积所需的训练数据量从超过4000例患者减少到不到220例。该性能上级改良早期预警评分（AUROC：0.76），并且对应于临床数据收集时间从约6个月缩短至不到10天。我们的研究结果突出了迁移学习在CDS系统专业化到新医院站点中的有用性，而不需要昂贵和耗时的数据收集工作。
Algorithm–based clinical decision support (CDS) systems associate patient-derived health data with outcomes of interest, such as in-hospital mortality. However, the quality of such associations often depends on the availability of site-specific training data. Without sufficient quantities of data, the underlying statistical apparatus cannot differentiate useful patterns from noise and, as a result, may underperform. This initial training data burden limits the widespread, out-of-the-box, use of machine learning–based risk scoring systems. In this study, we implement a statistical transfer learning technique, which uses a large “source” data set to drastically reduce the amount of data needed to perform well on a “target” site for which training data are scarce. We test this transfer technique with AutoTriage, a mortality prediction algorithm, on patient charts from the Beth Israel Deaconess Medical Center (the source) and a population of 48 249 adult inpatients from University of California San Francisco Medical Center (the target institution). We find that the amount of training data required to surpass 0.80 area under the receiver operating characteristic (AUROC) on the target set decreases from more than 4000 patients to fewer than 220. This performance is superior to the Modified Early Warning Score (AUROC: 0.76) and corresponds to a decrease in clinical data collection time from approximately 6 months to less than 10 days. Our results highlight the usefulness of transfer learning in the specialization of CDS systems to new hospital sites, without requiring expensive and time-consuming data collection efforts.