Demonstration of the Feasibility of Closed Loop Control of Blood Glucose in the Intensive Care Unit Setting Using a Novel Artificial Intelligence Based Glucose Control System

使用基于人工智能的新型血糖控制系统演示重症监护病房中血糖闭环控制的可行性

• 批准号: 10631190
• 负责人: Leon DeJournett
• 金额: $ 93.11万
• 依托单位: IDEAL MEDICAL TECHNOLOGIES, INC.
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-06-01 至 2025-05-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10631190
• 关键词: Address; Admission activity; Adult; Algorithms; Artificial Intelligence; Artificial Pancreas; Blood Glucose; COVID-19 pandemic; Clinical; Clinical Trials; Collaborations; Computer software; Continuous Glucose Monitor; Critical Illness; Data; Development; Devices; Discipline of Nursing; Dose; Engineering; Environment; Failure; Food Packaging; Glucose; Grant; Head; Hour; Human; Hyperglycemia; Hypoglycemia; Infusion procedures; Insulin; Intensive Care Units; Intravenous; Legal patent; Link; Liver; Manuals; Marketing; Measures; Medical Device; Medical Technology; Methods; Monitor; Morbidity - disease rate; Names; Nurses; Pancreas; Patient Admission; Patient Care; Patient-Focused Outcomes; Patients; Pattern; Persons; Phase; Phase III Clinical Trials; Physicians; Process; Programming Languages; Pump; Randomized; Randomized, Controlled Trials; Reading; Resistance; Risk; Safety; Speed; Stress; System; Technology; Testing; Time; Translating; Treatment Protocols; United States; United States Food and Drug Administration; Universities; Variant; artificial intelligence algorithm; blood glucose regulation; commercialization; cyber security; design; diabetic; effectiveness evaluation; efficacy study; efficacy trial; experience; first-in-human; glucose monitor; human error; improved; in silico; in vivo; innovation; instrument; mortality; non-diabetic; novel; phase III trial; porcine model; prospective; safety study; usability

PROJECT SUMMARY Failure to achieve tight glucose control in intensive care unit (ICU) patients is a major cause of morbidity and mortality. The current method for controlling glucose for the almost 6 million patients admitted to the ICU each year consists of time-intensive, error-prone, manual processes in which nurses have to check patient glucose levels, calculate dosing and rates for intravenous (IV) insulin, and adjust IV pumps. These steps can decrease nursing efficiency by up to 2 hours per patient per day. Ideal Medical Technologies is developing an improved system for glucose control in ICU patients. FUSIONTM, our closed-loop glucose control system, acts as an artificial pancreas; our innovation centers on using biomimicry to design our artificial intelligence-based algorithms so that they function like the glucose regulation system natively present in humans. These algorithms allow FUSION to continuously monitor a patient’s blood glucose level and autonomously infuse insulin and/or dextrose into the patient to maintain the blood glucose level within a physician-prescribed range. An automated, closed-loop approach to regulating glucose in critically ill patients would represent a paradigm shift in treatment protocols, and ultimately, morbidity and mortality rates in the ICU. One potential additional benefit of a closed-loop glucose control system beyond improved patient outcomes is the possibility of decreasing nursing time spent on glucose control. The FDA designated FUSION as a Breakthrough Medical Device in 2019, and our original IDE application was approved by the FDA as of this year. In this proposed project, we intend to optimize a commercial FUSION system and test it first in a 10-person non-randomized clinical safety trial and then in a 30-person randomized controlled safety and efficacy trial – both trials will be performed in an ICU at Emory University. Successful completion of this project will demonstrate clinical safety and preliminary efficacy of the FUSION closed-loop glucose control system and will enable us to then move into larger efficacy Phase III trials in a subsequent stage of development and commercialization; these Phase III trials will complete our data package for the FDA for marketing approval.

项目摘要 在重症监护病房（ICU）患者中未能实现紧密的葡萄糖控制是发病率和 死亡。当前用于控制葡萄糖的近600万患者的方法 年份由时间密集型，容易出错的手动过程组成，其中护士必须检查患者葡萄糖 水平，计算静脉（IV）胰岛素的剂量和速率，并调整IV泵。这些步骤可以减少 护理效率每天最多每天2个小时。 理想的医疗技术正在开发一种改进的ICU患者葡萄糖控制的系统。 FusionTM， 我们的闭环葡萄糖控制系统充当人造胰腺；我们的创新以使用仿生为中心 设计基于人工智能的算法，以便它们像葡萄糖调节系统一样起作用 本人在人类中存在。这些算法允许融合不断监测患者的血糖 水平并自主注入胰岛素和/或葡聚糖，以维持血糖水平 物理处方范围。一种自动化的闭环方法，用于调节重症患者的葡萄糖 将代表治疗方案的范式转移，最终是ICU中的发病率和死亡率。 闭环葡萄糖控制系统的一个潜在额外好处超出了改善的患者预后 减少护士在葡萄糖控制上花费的时间的可能性。 FDA指定为2019年的突破性医疗设备，我们的原始IDE应用程序是 截至今年，获得FDA的批准。在这个拟议的项目中，我们打算优化商业融合 系统并在10人的非随机临床安全试验中首先对其进行测试，然后在30人的随机中进行测试 受控的安全性和效率试验 - 这两个试验将在埃默里大学的ICU进行。 该项目的成功完成将证明融合的临床安全性和初步效率 闭环葡萄糖控制系统，将使我们能够进入更大的效率III期试验 随后的发展和商业化阶段；这些第三阶段试验将完成我们的数据包 用于营销批准的FDA。