Calibrating predictive model estimates to support personalized medicine.
Calibrating predictive model estimates to support personalized medicine.
复制标题
DOI：
10.1136/amiajnl-2011-000291
复制
发表时间：
2012-03
期刊：
影响因子：
--
通讯作者：
Ohno-Machado L
中科院分区：
文献类型：
--
作者：
Jiang X;Osl M;Kim J;Ohno-Machado L
为医学相关结果生成个性化估计的预测模型在临床护理和转化研究中发挥着越来越重要的作用。然而，目前用于校准这些估计值的方法丢失了有价值的信息。我们的目标是开发一种新的校准方法，以保存尽可能多的信息，并会比较有利的重要性能指标方面的现有方法：歧视和校准。我们提出了一种自适应技术，利用个性化的置信区间（CI）来校准预测。我们评估这种新方法，自适应校准的预测（ACP），在人工和现实世界的医疗分类问题，在ROC曲线下的区域，Hosmer-Lemeshow拟合优度检验，均方误差和计算复杂性。ACP优于其他校准方法，如分箱，普拉特缩放，和保序回归。在几个实验中，分箱，保序回归，普拉特缩放未能提高逻辑回归模型的校准，而ACP一贯提高校准，同时保持相同的歧视，甚至提高它在一些实验中。此外，ACP算法在计算上并不昂贵。对于某些预测模型，计算单个预测的CI可能很麻烦。ACP并非完全无参数：CI的长度可能会影响其结果。ACP可以生成可能比使用现有方法校准的估计更适合于个性化预测的估计。需要进一步研究以探索ACP的局限性。
Predictive models that generate individualized estimates for medically relevant outcomes are playing increasing roles in clinical care and translational research. However, current methods for calibrating these estimates lose valuable information. Our goal is to develop a new calibration method to conserve as much information as possible, and would compare favorably to existing methods in terms of important performance measures: discrimination and calibration. We propose an adaptive technique that utilizes individualized confidence intervals (CIs) to calibrate predictions. We evaluate this new method, adaptive calibration of predictions (ACP), in artificial and real-world medical classification problems, in terms of areas under the ROC curves, the Hosmer-Lemeshow goodness-of-fit test, mean squared error, and computational complexity. ACP compared favorably to other calibration methods such as binning, Platt scaling, and isotonic regression. In several experiments, binning, isotonic regression, and Platt scaling failed to improve the calibration of a logistic regression model, whereas ACP consistently improved the calibration while maintaining the same discrimination or even improving it in some experiments. In addition, the ACP algorithm is not computationally expensive. The calculation of CIs for individual predictions may be cumbersome for certain predictive models. ACP is not completely parameter-free: the length of the CI employed may affect its results. ACP can generate estimates that may be more suitable for individualized predictions than estimates that are calibrated using existing methods. Further studies are necessary to explore the limitations of ACP.
登录
查看更多内容
DOI：
10.1038/nrcardio.2010.53
发表时间：
2010-06
期刊：
Nature reviews. Cardiology
影响因子：
--
作者：
通讯作者：
--
影响因子：
5
作者：
Karp, I;Abrahamowicz, M;Pilote, L
通讯作者：
Pilote, L
影响因子：
1.1
作者：
Hooks, Mary A.
通讯作者：
Hooks, Mary A.
DOI：
10.1136/amiajnl-2011-000245
发表时间：
2011-07-01
影响因子：
6.4
作者：
Sarkar, Indra Neil;Butte, Atul J.;Ohno-Machado, Lucila
通讯作者：
Ohno-Machado, Lucila
影响因子：
6.2
作者：
Ayer, Turgay;Alagoz, Oguzhan;Chhatwal, Jagpreet;Shavlik, Jude W.;Kahn, Charles E., Jr.;Burnside, Elizabeth S.
通讯作者：
Burnside, Elizabeth S.