INTELLIGENT REAL-TIME INFORMATION SYNTHESIS FOR MONITORS

综合来自多个实时数据源的信息， 对患者的潜在疾病进行智能和连贯的评估， 血流动力学状况对发展至关重要 智能医疗监护仪 Dean Sittig博士目前正在指导 智能心血管监护仪（ICM）原型的开发。 的 建议的项目建立在这项工作，包括：1）确定 多个实时数据中具有医学意义的趋势和伪影 源，2）生成患者特定的“智能”警报，以及3）测试 和改进的算法，旨在识别几个基本的 血液动力学异常（例如，血容量不足或心脏压塞等）。 这项拟议的研究项目将特别侧重于智能 从多个实时数据源导出的信息的合成 (i.e.,趋势检测、伪影识别和生理状态 确定算法），以生成对 患者的血液动力学状态。 最后，我们将使用来自 重症监护室和手术室。 趋势和人工制品 检测到、生成报警和识别血流动力学异常将 将其与“金标准”注释的医疗记录进行比较， 临床工作人员，以帮助评估监护仪在以下方面的效用 上面列出的目标。

Synthesizing information from multiple real-time data sources to generate an intelligent and coherent assessment of a patient's underlying hemodynamic condition is of vital importance for the development of intelligent medical monitors. Dr. Dean Sittig is currently directing the development of a prototype intelligent cardiovascular monitor (ICM). The proposed project builds on this work and includes: 1) identifying medically meaningful trends and artifacts from multiple real-time data sources, 2) generating patient-specific "smart" alarms, and 3) testing and refinement of algorithms designed to recognize several basic hemodynamic abnormalities (e.g., hypovolemia or cardiac tamponade, etc.). This proposed research project will focus particularly on the intelligent synthesis of information derived from multiple real-time data sources (i.e., the trend detection, artifact recognition, and physiological state determination algorithms) to generate a more systematic assessment of a patient's hemodynamic status. Finally, we will test a prototype of the ICM using data from the intensive care unit and the operating room. Trends and artifacts detected, alarms generated, and hemodynamic abnormalities recognized will be compared with a "gold standard" annotated medical record generated by the clinical staff, to help assess the utility of the monitor in terms of the goals listed above.