A closed-loop gravity infusion control device

项目摘要 这项工作将开发一种创新的输液设备，利用新的电子和传感器 生产改进静脉内（IV）输注输送的医疗装置的能力。IV 输注是世界范围内最常见的医疗保健程序之一; 据估计，80%的住院患者接受静脉注射治疗。输液也越来越多 在医院环境之外、家庭、长期护理机构、激增环境和 其他备用现场护理（ASC）环境。特别是ASC输注， 医疗保健趋势，因为它们通过消除障碍（如患者 运输、免疫功能低下者的病原体暴露和住院费用。而 输液是医疗保健服务的关键基础设施，提供输液的技术还没有 跟上行业不断变化的需求。该建议建立在临床验证的基础上， 输液监测系统，即DripAssist输液速率监测器，以创建闭环， 能够施用甚至可能复杂的药物的数字连接的输注平台 在急性和ASC环境中。这项技术支持的重力输注输送 具有自动流量调节的系统，我们称之为DripAssist控制，将使用低成本 传感器来精确测量液滴中的液体体积。这是一项实验研究， 将通过计算机收集预定的因变量和自变量的数据。 大型数据集将被收集，传统的统计分析和机器 学习方法将得到应用。我们的研究将利用丰富的信息证实， 在现有初步台式系统的输出中呈现，以建立新的数据处理 测量液滴体积的方法。一旦完善，这些方法将被优化为 闭环流量控制系统，将其开发成一个形式和适合的设备和测试 在具有现实约束和可变性的临床环境中。我们有三个目标 工作目标1：建立两个能够广泛识别滴液量的台式系统 通过新颖的检测过程对输液器进行评级。目标2：优化我们的传感器， 将算法驱动到临床上可接受的精度和准确度。目标3：执行闭环 控制系统集成和临床前开发工作，以获得工作原型 可以测试可用性和已建立的监管标准的器械。我们相信我们的 这项工作可以通过引入新的仪器， 传感器和生产更简单的输液输送装置， 昂贵的泵平台的精度。

Project Summary This work will develop an innovative infusion device that leverages new electronics and sensor capabilities to produce a medical device that improves intravenous (IV) infusion delivery. IV infusions are among most common healthcare procedures administered worldwide; an estimated 80% of hospital patients receive some IV therapy. Infusions are also increasingly delivered outside of hospital settings, in homes, long-term care facilities, surge settings and other alternate site care (ASC) environments. ASC infusions, in particular, are a critical healthcare trend because they increase access to care by eliminating obstacles such as patient transportation, pathogen exposure for the immunocompromised, and inpatient costs. While infusion is critical infrastructure for healthcare delivery, the technologies to deliver it have not kept up with changing needs in the sector. This proposal builds on a clinically-validated infusion monitoring system, the DripAssist Infusion Rate Monitor, to create a closed-loop, digitally-connected infusion platform capable of administering even potentially complex drug regimes in both acute and ASC settings. This technology-enabled gravity infusion delivery system with automatic flow adjustment, what we call DripAssist Control, will use low cost sensors to accurately measure the volume of fluid in a drop. This is an experimental study that will collect data in silico across a predetermined set of dependent and independent variables. Large data sets will be collected to which both conventional statistical analysis and machine learning methods will be applied. Our research will utilize rich information confirmed to be present in outputs of an existing preliminary benchtop system to establish new data processing methods for measuring drop volume. Once refined, these methods will be optimized into a closed-loop flow control system, which will be developed into a form and fit device and tested within clinical settings with realistic constraints and variability. We have three Aims in this work. Aim 1: establishing two benchtop systems that can broadly identify the drip volume rating of an infusion set through a novel detection process. Aim 2: optimize our sensing and actuating algorithms to clinically acceptable precision and accuracy. Aim 3: perform closed-loop control system integration and preclinical development work to attain a working prototype device that can be tested for usability and established regulatory standards. We believe our work can create a bold shift in clinical practice by introducing novel instrumentation and sensors and producing a simpler infusion delivery device that matches the precision and accuracy of expensive pump platforms.