A pathophysiology driven spatial dynamic modeling framework for personalized prediction and precision medicine

摘要 预测性个性化医疗和精准医疗使医学进入了一个新时代，在这个时代，传统的 生理信息和新的临床数据，包括遗传数据、成像数据和健康数据 共同为正确的患者在正确的时间进行正确的治疗。医疗保健和临床 随着科技的快速发展，海量、多种多样的临床数据变得可广泛获取和使用。 然而，目前的建模技术缺乏将病理生理学和成像数据相结合。因此， 开发一种创新的数据分析工具，将个性化临床数据与现有科学相结合 流行病学和临床医学成为这一新领域的迫切需要。它的总体愿景是 研究项目是开发一种病理生理学驱动的空间动态建模(PDSDM)方法 个性化医疗预测和精准医疗。我们的五年目标是发展PDSDM 计算建模平台，并在各种疾病的患者数据上验证该平台。具体来说，我们 我将开发一个交互式计算平台来构建PDSDM模型，并开发一个 模块自动模拟模型。然后，我们将开发一个模型校准模块，使用 将生理信号通路产生的数学模型参数化的临床患者数据 网络，还将把成像数据合并为空间计算域；此外，最佳 个性化治疗研究将在此计算建模平台上进行，以满足当前可用的 临床试验。这一创新的框架将整合数学建模、计算方法、数据 分析和数据驱动的优化技术，以提供个性化的空间计算模型 每一个人。我们将在心血管等各种生物医学疾病上验证这一新框架 疾病、慢性胰腺炎和阿尔茨海默病，以及现有的临床和生物学数据。建议数 研究意义重大，因为它将为个性化疾病进展提供3D预测， 将更准确地评估个性化的疾病风险。它还将提供一种系统的方法来评估 可用的治疗计划几乎可以为每个人提供最佳的治疗建议。

Summary Predictive personalized healthcare and precision medicine enable a new era of medicine, in which traditional physiological information and new clinical data including genetic data, imaging data, and healthcare data come together to match the right patient with the right treatment at the right time. With healthcare and clinical technology rapidly grow, the vast and varied amounts of clinical data become widely accessible and usable. However, current modeling techniques lack of combining pathophysiology with imaging data. Therefore, developing an innovative data analytical tool that combines personalized clinical data with the existing sciences of epidemiology and clinical medicine becomes an urgent need in this new area. The overall vision of this research project is to develop a pathophysiology driven spatial dynamic modeling (PDSDM) approach for personalized healthcare prediction and precision medicine. Our five-year goals are to develop the PDSDM computational modeling platform and validate this platform on patient data for various diseases. In specific, we will develop an interactive computational platform to build the PDSDM model and develop a computational module to simulate the model automatically. Then we will develop a model calibration module by employing the clinical patient data to parameterize mathematical models arising from physiological signaling pathway networks and will also incorporate the imaging data as the spatial computational domain; moreover, optimal personalized treatment studies will be performed on this computational modeling platform for current available clinical trials. This innovative framework will integrate mathematical modeling, computational methods, data analysis, and data-driven optimization techniques to provide a personalized spatial computational model for each individual. We will validate this new framework on various biomedical diseases such as cardiovascular disease, chronic pancreatitis, and Alzheimer’s disease with existing clinical and biological data. The proposed research is significant because it will provide the 3D prediction for personalized disease progression which would evaluate personalized disease risk more accurately. It will also provide a systematic way to assess the available treatment plans virtually then to provide an optimal treatment suggestion for each individual.