Development of predictive analytics and business intelligence tools for improved patient management using intelligent decision support systems

该项目的目的是开发预测分析和商业智能工具，这些工具可以集成到医疗保健合作伙伴正在开发的决策支持系统中；Alpha Global IT作为其旗舰电子病历Globemed的一个模块。Alpha Global IT致力于开发先进的医疗保健IT解决方案，利用预测分析和商业智能来提高全省和全国患者的医疗保健质量。预测分析和建模工具将用于对人群进行“分层”，识别高危患者，提供更好的供应筛选测试，并确定护理差距。在这项特别的研究中，研究团队解决了医疗保健合作伙伴感兴趣的以下任务：(1)开发预测分析和建模工具，利用贝叶斯网络来预测医疗保健系统在癌症新发病率方面的负荷。因此，这些预测模型可以用来洞察医疗保健系统所需的资源，以便有效地管理预计的癌症发病率并提供所需的护理水平。(2)通过引入进化计算方法扩展Alpha Global IT的非线性优化工具，从而创建一个调度工具，以改善临床资源的分配。此工具可用于实现医疗保健组织的性能目标（所需的流程数量、等待时间要求、地理位置等）。在护理点需要商业智能工具，因此医生可以快速解释数据，定义“健康干预”，并支持诸如专家转诊和疾病管理等操作。在这方面，该研究项目计划引入新的工具，用于从临床记录中提取适应性信息，以帮助创建最佳实践的证据。这种自适应信息提取工具将成为医疗保健合作伙伴引入的临床决策支持系统的一个关键特征。

The purpose of this project is to develop predictive analytics and business intelligence tools that can be integrated into the decision support system being developed by the healthcare partner; Alpha Global IT as a module of the their flagship electronic Medical Record, Globemed. Alpha Global IT is committed to the development of advanced healthcare IT solutions that utilize predictive analysis and business intelligence in order to improve the quality of healthcare for patients within the Province and Nationwide. Predictive analytics and modeling tools will be used to "stratify" the population, identify at-risk patients, provide better provision screening tests, and identify gaps in care. In this particular research, the research team tackle the following tasks that are of interest to the Healthcare partner: (1) developing predictive analytics and modeling tools that utilize Bayesian networks to forecast the load on the healthcare system in terms of the projections for new incidences of cancer. These predictive models can hence be used to provide insight into the resources needed by the healthcare system in order to effectively manage the projected cancer occurrences and provide the level of care needed, (2) extend Alpha Global IT's nonlinear optimization tools through the introduction evolutionary computation methods in order to create a scheduling tool for improved allocation of clinical resources. This tool can be used to achieve healthcare organization performance targets (required number of procedures, wait time requirements, geographic locations, etc.). Business intelligence tools are needed at the point of care, so physicians can quickly interpret data, define "health interventions," and support actions such as specialist referrals, and disease management. On this front, this research program plans to introduce novel tools for adaptive information extraction from clinical records to help create evidence for best practice. Such adaptive information extraction tools will be a key feature of the clinical decision support system introduced by the healthcare partner.