SBIR Phase I: Development of novel artificial intelligence (AI)-enabled, non-invasive, heart attack diagnostics

这项小企业创新研究（SBIR）第一阶段项目的广泛影响是一种新型的便携式诊断设备，用于实时诊断心肌梗死（MI）和缺血，具有高水平的灵敏度和特异性。该系统旨在通过实时诊断和监测工具，为美国每年发生的80.5万例心脏病发作病例提供准确的即时临床分类。该系统旨在显著减少当前侵入性采样和血液分析措施所需的时间，从而在心肌梗死的头几个关键小时内改善患者的预后，同时节省医疗资源并提高效率。该技术的便携性使其他形式的集成成为可能，包括在家庭、诊所、医院床边使用或现场使用应用。这个小企业创新研究（SBIR）第一阶段项目将开发一个基于概念验证的诊断预测机器学习系统，用于检测MI。活动范围包括测试多种技术和模型，生成分布分析和训练数据的事件图，以便与临床判定的事件相比优化性能。成功的措施包括一个数据模型，当使用他们专有的外部无创透皮生物标志物传感器和他们的可穿戴服装心电图（ekg）时，能够以85%的准确率检测心肌梗死。该设备还能够以85%的准确率检测临床相关的缺血。这些结果将推动公司完成独立、无创、实时诊断和远程监控工具的目标。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。

The broader impact of this Small Business Innovation Research (SBIR) Phase I project is a novel, portable, diagnostic device for non-invasively diagnosing myocardial infarctions (MI) and ischemia in real-time with a high level of sensitivity and specificity. The system aims to provide an accurate point-of-care clinical classification for the 805,000 heart attack cases occurring in the US each year with a real-time diagnostic and monitoring tool. The system aims to significantly reduce the time needed with current invasive sampling and blood analysis measures, thereby improving patient outcomes during the first critical hours of an MI while saving healthcare resources and improving efficiency. The portable nature of the technology enables other forms of integration including at home, in clinic, hospital bedside use, or field use applications.This Small Business Innovation Research (SBIR) Phase I project will develop a proof-of-concept diagnostic-prognostic machine learning-based system for detecting MI. The scope of activities includes testing multiple techniques and models and producing distribution analyses and event plots of training data in order to optimize performance compared to clinically adjudicated events. Success measures include a data model which, when using their proprietary external noninvasive transdermal biomarker sensor with their wearable garment electrocardiograms (EKGs), are able to detect MI with 85% accuracy. The device will also be able to detect clinically relevant ischemia with 85% accuracy. These results will progress the company’s objective of completing a standalone, noninvasive, real-time diagnostic and remote monitoring tool.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.