Non-contact In-Bed Sensing of Heart Failure Decompensation

项目摘要： 慢性心力衰竭是一种高度流行和代价高昂的疾病，其特点是经常恶化。 导致反复住院治疗。心力衰竭患者有非常高的 体液过多的住院率，如果检测到病情恶化，通常可以避免 在门诊加强早期和利尿治疗。不幸的是，自我- 体重测量和症状自我报告已被证明无效。的效用 家庭监控技术一直受到自我测量、自我测量和自我监控的不良影响 数据传输，或可穿戴设备的自我应用。植入型设备取得了成功 但都是侵入性的，而且成本高昂。非接触式传感器目前依赖于位置，不 兼容睡在同一张床上的多人。南丁格尔实验室的成立是为了 将一种名为BedScale的非接触式黏附独立传感器商业化。我们的预赛 结果显示，这种低调的机械传感器，当放置在家中的腿下时 床上，可以执行粘着独立的总体重，运动， 呼吸和心脏信号长达数月而没有患者互动，即使共用一张床 被伴侣或宠物。此第一阶段SBIR的总体目标是以真实的- 世界心力衰竭人群正在为第二阶段的前瞻性临床研究做准备。 定义即将住院的签名，BedScales旨在使门诊 心力衰竭的管理是一门以数据为导向的科学。

PROJECT ABSTRACT: Chronic heart failure is a highly prevalent and costly disease marked by frequent exacerbations leading to recurrent hospitalizations. Heart failure patients have extraordinarily high hospitalization rates for fluid overload, which can often be avoided if exacerbations are detected early and diuretic therapy is intensified in the outpatient setting. Unfortunately, self- measurement of weights and self-reporting of symptoms have proven ineffective. The utility of home monitoring technologies has been limited by poor adherence to self-measurement, self- transmission of data, or self-application of wearables. Implantable devices have shown success but are invasive and expensive. Non-contact sensors are currently position-dependent and not compatible with multiple persons sleeping in the same bed. Nightingale Labs was founded to commercialize a non-contact adherence-independent sensor called Bedscales. Our preliminary results show that the low-profile mechanical sensors, when placed beneath the legs of a home bed, can perform adherence-independent sensing of total body weight, movements, respirations, and cardiac signals for months without patient interaction, even if the bed is shared by partners or pets. The overall goal of this Phase I SBIR is to define sensor accuracy in a real- world heart failure population in preparation for a prospective clinical study in Phase 2. By defining the signatures of impending hospitalization, BedScales aims to make outpatient management of heart failure a data-driven science.