SenSE: Closed-loop Artificial Pancreas with Noninvasive Monitoring of Glucose and Diet

摘要题目：意义：具有无创血糖和饮食监测功能的人工胰腺。项目编号：2037383。主要研究人员：Xia Zhou， Gregory Forlenza, Liu Qiang, Temiloluwa Prioleau & Tam vuu。研究机构：达特茅斯学院，科罗拉多大学和德克萨斯大学。大多数1型患者不能成功地达到目标血糖水平，尽管近年来技术的应用越来越广泛，但血糖控制实际上一直在恶化。控制不当的糖尿病是失明、肾衰竭、心脏病发作和周围血管疾病的主要原因，可能导致截肢，在某些情况下甚至死亡。联合使用连续血糖监测仪和胰岛素泵被认为是目前治疗1型糖尿病的黄金标准。然而，现有的技术存在两个既定的局限性：由于需要手动“膳食通知”来估计胰岛素剂量，用户负担高；由于不断打破皮肤以插入传感器的不便和传感器粘合剂的刺激，葡萄糖监测设备的粘附性低。因此，在公共卫生领域迫切需要对1型糖尿病患者进行全自动闭环血糖控制，这将是本次研究的重点。技术摘要：提出的工作旨在开发和试点一种联合多模态传感系统，该系统除了可以估计膳食摄入量的自动膳食检测外，还可以提供连续的无创血糖监测数据。这两种类型的信息都是人工胰腺算法预测大剂量（即短效）胰岛素需求所必需的。更具体地说，所提出的系统采用耳戴式设备的形式，解决了实现全闭环人工胰腺系统的关键系统和算法挑战。以下是所提议的工作在智力上的优点。首先，它将开发新的系统设计，通过融合三种信号流（即生物电信号、运动和声学信号）来实现细粒度、鲁棒的饮食传感；其次，它将通过解决皮肤散射的挑战，减轻混杂因素和用户多样性的影响，推进现有的基于光学的无创血糖监测；第三，它将提出新的节能机器学习模型，致力于在微控制器上实现本地执行，处理有限数量的标记训练数据，并确保数据隐私和通信效率；最后，它将表征和验证与糖尿病管理最相关的生物和行为生物标志物，以实施人工胰腺系统，并开发自适应采样和噪声处理的新方法，以确保准确监测目标事件。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Abstract Title:SenSE: Closed-loop Artificial Pancreas with Noninvasive Monitoring of Glucose and DietProposal Number: 2037383Principal Investigator: Xia Zhou, Gregory Forlenza, Qiang Liu, Temiloluwa Prioleau & Tam VuInstitutions: Dartmouth College, University of Colorado & University of TexasNon-technical Abstract:Type-1 diabetes affects over 1.25 million people in the United States and its incidence is increasing every year. The majority of type 1 patients are not successful in achieving target blood glucose levels and control has actually been worsening in recent years despite expanded use of technologies. Inadequately controlled diabetes is the leading cause of blindness, kidney failure, heart attack and peripheral vascular disease possibly leading to amputation and in some cases death. Joint use of a continuous glucose monitor and insulin pump is considered the current gold standard for management of type-1 diabetes. Existing technologies, however, suffer from two well-established limitations: high user burden due to the need for manual “meal announcements” to estimate insulin doses and low adherence to glucose monitoring devices due to the inconvenience of constantly breaking the skin for sensor insertion and irritation from the sensor adhesive. Thus, there is an urgent need in public health for fully automated, closed-loop glycemic control for type-1 diabetes patients, which will be the focus of the proposed research.Technical Abstract:The proposed work aims to develop and pilot a joint multimodal sensing system that can provide continuous noninvasive glucose monitoring data in addition to automated meal detection with an estimate of meal intake. Both types of information are necessary to inform artificial pancreas algorithms for prediction of bolus (i.e. short-acting) insulin dose needs. More specifically, the proposed system is in the form of an ear-worn device, tackling key systems and algorithmic challenges to realize a fully closed-loop artificial pancreas system. The intellectual merits of the proposed work are following. First, it will develop novel system designs to realize fine-grained, robust dietary sensing by fusing three streams of signals (i.e., bio-electrical signals, motion, and acoustic signals); Second, it will advance existing optical-based noninvasive glucose monitoring by addressing challenges of skin scattering and mitigating the impact of confounding factors and user diversity; Third, it will present new energy-efficient machine learning models dedicated to enabling local execution at micro-controllers, dealing with the limited number of labeled training data, and ensuring data privacy and communication efficiency; Finally, it will characterize and validate biological and behavioral biomarkers most relevant to diabetes management for implementing an artificial pancreas systems and develop new approaches for adaptive sampling and noise handling to ensure accurate monitoring of target events.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.