Efficient Statistical Learning Methods for Personalized Medicine Using Large Scale Biomedical Data

项目摘要： 冠状病毒病19（COVID-19）在世界各地造成了重大的公共卫生危机。小说 冠状病毒的潜伏期长，在此期间极具传染性。没有经过验证的 有效的治疗方法或疫苗。许多国家实施了大规模的公共干预措施 和美国各州在疫情的不同阶段，有不同的社会疾病组合， 流动限制和人口行为变化。关于如何实施这些干预措施的决定 (e.g.,何时实施和放松缓解措施）依赖于COVID流行病学的重要统计数据（例如， 有效繁殖数）来表征和预测COVID-19爆发的过程。不过有 缺乏稳健和简约的COVID流行病模型，可以准确反映 易感人群和区域（例如，人口统计、医疗保健能力、社会和经济决定因素）。 没有严格的研究来指导针对某一人群或地区的精准公共卫生干预措施 这取决于它们的特性。此外，由于公共卫生干预的非随机性， 在比较不同地区的COVID-19缓解措施时， gions。为了应对这些挑战，该项目开发了强大的和可推广的分析方法来评估 公共卫生干预措施，并评估个体患者感染COVID-19和并发症的风险。在目标1中， 我们会发展动态和稳健的统计模式，以预测疾病的流行。模型将估计 第一个未知感染病例的日期，瞬时有效繁殖数，以及对感染的解释， COVID-19病毒的感染期。此外，人口统计、社会和经济方面的异质性 将纳入指标、医疗能力和地理位置，以反映其对COVID的影响 疫情在纵向准实验设计下，我们将为比较公众 在不同地区实施的卫生干预措施，同时考虑到混杂的偏见。多种来源的 来自美国不同州的数据将进行分析，以实证检验哪些州的应对策略更 有效性和在哪个亚群中。在目标2中，我们将专注于开发精确的个体风险评估工具， 使用在纽约纽约长老会医院收集的电子健康记录（EHR）的年度COVID-19患者 约克市，COVID-19的震中。我们将工程师的特点，病人的先决条件相关的严重 COVID并发症、康复或死亡。更重要的是，我们将设计代表病毒代理的功能， 患者的地理信息。我们将使用机器学习技术来创建定量的 患者预后的概要（例如，转化为严重临床阶段、出院、死亡）。我们将使用- 最终交叉验证和外部校准，以验证开发的算法。该项目将产生证据 指导精确的公共卫生干预，最佳的病人护理和有效的医疗资源分配， 预期第二波COVID疫情及为其他传染病爆发作好准备。

Project Summary: Coronavirus disease 19 (COVID-19) has created a major public health crisis around the world. The novel coronavirus was observed to have a long incubation period and extremely infectious during this period. No proven effective treatment or vaccine is available. Massive public interventions have been implemented in many countries and states in the United States (US) at different phases of the outbreak with varying combinations of social dis- tancing, mobility restriction and population behavioral change. Decisions on how to implement these interventions (e.g., when to impose and relax mitigation measures) rely on important statistics of COVID epidemiology (e.g., effective reproduction number) that characterize and predict the course of COVID-19 outbreak. However, there is a lack of robust and parsimonious model of COVID epidemic that can accurately reflect the heterogeneity between susceptible populations and regions (e.g., demographics, healthcare capacity, social and economic determinants). There is no rigorous study to guide precision public health interventions that are tailored to a population or region depending on their characteristics. Furthermore, due to the non-randomized nature of public health interventions, it is critical to account for biases and confounding when comparing mitigation measures of COVID-19 across re- gions. To address these challenges, this project develops robust and generalizable analytic methods to evaluate public health interventions and assess individual patient risks of COVID-19 infection and complications. In Aim 1, we will develop dynamic and robust statistical models to predict the disease epidemic. The models will estimate the date of the first unknown infection case, instantaneous effective reproduction number, and account for the incu- bation period of COVID-19 virus. Furthermore, heterogeneity in population's demographics, social and economic indicators, healthcare capacity and geographic locations will be incorporated to reflect their impacts on COVID epidemic. Under a longitudinal quasi-experimental design, we will provide valid inference for comparing public health interventions implemented at different regions while accounting for confounding bias. Multiple sources of data from different states in the US will be analyzed to empirically test which states' response strategies are more effective and in which subpopulation. In Aim 2, we will focus on developing precise risk assessment tool of individ- ual COVID-19 patients using electronic health records (EHRs) collected at New York Presbyterian hospital in New York City, an epicenter of COVID-19. We will engineer features of patient's pre-conditions associated with severe COVID complications, recovery, or death. More importantly, we will engineer features that represent proxies of virus exposures from patients' geographic information. We will use machine learning techniques to create quantitative summaries of patient prognosis (e.g., transitioning to serious clinical stages, discharge, death). We will use inter- nal cross-validation and external calibration to validate developed algorithms. The project will generate evidence to guide precision public health intervention, optimal patient care, and efficient healthcare resource allocation in anticipation of a second wave of COVID epidemic and in preparation of other infectious disease outbreaks.