SBIR Phase II: An AI-powered wearable system and platform for long-term remote monitoring of pulmonary function (COVID-19)

这个小型企业创新研究（SBIR）第二阶段项目的更广泛的影响/商业潜力是改善肺部疾病的护理，如慢性阻塞性肺疾病（COPD）。这种疾病影响了3000万美国人，每年造成15万多人死亡，是美国死亡的第三大原因。COPD每年花费美国人近720亿美元，其中一半用于急诊和因呼吸危机而住院。住院的COPD患者生活质量下降，肺部疾病进展更快，随后两年的死亡率增加50%。目前，COPD肺恶化的持续监测主要依赖于患者报告的症状和不充分的工具（准确率为65-70%）。因此，一半的COPD恶化仍未被发现。所提出的技术通过持续远程跟踪肺功能促进COPD发作的早期诊断和治疗，从而防止不必要的ER就诊和住院。该设备改善了对慢性阻塞性肺病或哮喘等长期疾病的护理，也改善了对那些相对快速发病的疾病的护理，如COVID-19。拟议的项目推进了使用声学共振测量肺空气量的设备的翻译。该项目开发了一个扩展的数据集，用于大规模分析和部署。该项目进一步将高级分类模型纳入分析工作流程，以跟踪和预测云计算环境中恶化的肺功能。该项目还将开发一个版本，运行在病人的移动终端，以实现实时反馈。该奖项反映了NSF的法定使命，并已被认为是值得通过评估使用基金会的智力价值和更广泛的影响审查标准的支持。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project is to improve care for pulmonary conditions, such as Chronic Obstructive Pulmonary Disease (COPD). This disease affects 30 million Americans, kills more than 150,000 a year and is the third leading cause of U.S. deaths. COPD costs Americans nearly $72 billion a year, half of which is spent on ER visits and hospital stays resulting from respiratory crises. Hospitalized COPD patients have reduced quality of life, more rapid lung disease progression, and a 50% increase in mortality in the subsequent two years. Currently, ongoing monitoring of COPD lung deterioration relies largely on patient-reported symptoms and inadequate tools (65-70% accurate). As a result, half of COPD deterioration remains undetected. The proposed technology facilitates early diagnosis and treatment of COPD attacks by continuous remote tracking of lung function, thereby preventing unnecessary ER visits and hospitalizations. This device improves care for long-term conditions like COPD or asthma, but also those with relatively quick onset, such as COVID-19.The proposed project advances translation of a device using acoustic resonance to measure lung air volume. The project develops an expanded data set for analysis and deployment at scale. The project further incorporates advanced classification models into the analysis workflow to track and predict deteriorating lung function in a cloud-computing environment. The project will also develop a version to run on a patient's mobile device to enable real-time feedback.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.