I-Corps: Wearable Cardiovascular Abnormality Detector

这个i-Corps项目的更广泛的影响/商业潜力是开发一种可穿戴技术，通过临床外监测及早发现疾病，使患有可疑心血管异常的个人受益。早期发现心血管疾病可以及时进行干预，包括药物治疗、外科手术和瓣膜置换术。拟议技术的不引人注目的性质可能有助于提供流畅和舒适的体验，鼓励患者遵守诊断过程。这种依从性的改善，加上饮食和生活方式的改变，可能会通过降低住院率和相关成本来改善社会健康结果。诊所外的监测可能会消除后续临床访问的需要，将诊断时间从几周缩短到几天，并减少心脏病专家的工作量。这一i-Corps项目基于开发一种基于处方的可穿戴传感器贴片，用于对心血管患者进行长达两周的持续评估。该传感器贴片利用麦克风和惯性测量单元(即加速计和陀螺仪)来分别捕获心跳引起的声音和胸壁上的振动。开发了一种基于人工智能(AI)的软件工具，用于向心脏科医生通知心脏信号中的异常事件和模式及其相应的时间戳。该技术还提供血液动力学功能，如收缩和舒张期时间间隔和心率变异性参数，以帮助心脏病专家进行决策过程。该设备替代了传统的Holter监护仪和单导联心电图仪传感器，这些传感器不方便长时间佩戴，也不能代表心脏的机械活动。基于机器学习(ML)的心脏-机械和心声模式融合方法也可能适用于其他生物机械和生物-声学监测应用，如关节和骨骼健康监测，在这些应用中，异常的克隆氏症可能是关节功能障碍的指标。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this I-Corps project is the development of a wearable technology to benefit individuals with sustpected cardiovascular abnormalities by early detection of disease through outside-the-clinic monitoring. Early detection of cardiovascular diseases allows for timely interventions including medication, surgical procedures, and valve replacement. The unobtrusive nature of the proposed technology may contribute to a smooth and comfortable experience, encouraging patients to comply with the diagnostic process. This improved compliance, along with dietary and lifestyle modifications, may lead to improvements in societal health outcomes by decreasing hospitalization rates and the associated costs. Outside-the-clinic monitoring may obviate the need for subsequent clinical visits, shortening the diagnostic timeline from weeks to days and reducing cardiologists’ workload.This I-Corps project is based on the development of a prescription-based wearable sensor patch for the continuous assessment of cardiovascular patients for periods of up to two weeks. This sensor patch leverages microphones and inertial measurement units (i.e. accelerometers and gyroscopes) to capture heartbeat-induced sounds and vibrations on the chest wall, respectively. A software tool based on artificial intelligence (AI) is developed to inform cardiologists of abnormal events and patterns in the cardiac signals along with their corresponding timestamps. The technology also offers hemodynamic features such as systolic and diastolic time intervals and heart rate variability parameters to assist cardiologists with the decision-making process. This device is an alternative to traditional Holter monitors and single-lead electrocardiography (ECG) sensors that are inconvenient to wear for extended periods of time and do not represent the mechanical activities of the heart. The machine learning (ML)-based fusion method designed for cardio-mechanical and cardio-acoustic modalities may also be adapted to other bio-mechanical and bio-acoustic monitoring applications such as joint and bone health monitoring, where abnormal crepitus might be indicators of joint malfunctions.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.