SCC-PG: Just in Time Intervention for Patients with Chronic Heart Diseases in Arizona tribes

在美国，心血管疾病是导致死亡的主要原因。心血管疾病通常是慢性疾病，患者需要多次昂贵的急诊或长期住院。农村社区获得医疗设施的机会有限，可能导致这些患者，特别是生活在偏远和农村地区的美洲印第安人患者的健康状况恶化。相当多患有心血管疾病的亚利桑那州人工智能患者可能会因为远离医疗服务提供者而面临错过有效治疗机会窗口和生存机会较低的风险。亚利桑那州有第三大人口的印第安人，他们生活在农村，部落和通常非常孤立的地区。居住在这些地区的心脏病患者无法及时获得所需的护理，特别是心脏病专家等专业服务。因此，对农村患者而言，与这些情况相关的一项重要挑战是“早期发现症状恶化”，这对“及时”干预至关重要。该规划项目是北亚利桑那大学（NAU）和密歇根大学（UM）以及农村和部落健康社区领导人的合作成果，旨在讨论利用综合远程心脏监测系统的最佳策略，以使生活在农村，偏远和孤立部落地区的慢性心脏病患者受益。这一规划项目提供了若干创新方法，与部落人工智能社区合作，开始(i)开发一种新的远程心脏监测技术，以预测心房颤动和充血心力衰竭等一些常见慢性心脏病患者的症状恶化和严重心脏病的发生。一些远程心脏监测系统专注于对此类事件的可靠检测，然而，当设备提醒患者或其家人时，对于生活在农村和偏远地区的患者来说，寻求医疗帮助已经太晚了。因此，我们的早期预测框架可以给患者足够的时间寻求医疗援助。该系统还可以帮助护理人员控制严重症状，减少再入院，并降低与护理相关的成本；（ii）开发基于深度学习和马尔可夫的预测方法；（iii）开发可独立于云的设备上预测方法，以便为无法接入宽带互联网的患者提供服务。这项研究可以在广泛的其他疾病和医疗条件下以及在不同的地理区域进行复制。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。

Cardiovascular diseases are the leading cause of death in the United States. Cardiovascular diseases are often chronic conditions that involve several costly emergency visits or long hospitalizations for the patients. Limited access to medical facilities in rural communities can result in worse health outcomes for these patients, in particular, the American Indian (AI) patients living in remote and rural areas. A considerable number of Arizonan AIs with cardiovascular conditions may be at risk of missing the window of opportunity for effective treatment and experiencing a lower chance of survival because of living far away from medical service providers. Arizona has the third largest population of Indian Americans who live in rural, tribal and often extremely isolated areas. The cardiac patients living in these areas do not have the required timely access to care, in particular to specialty services like cardiologists. Therefore, an important challenge related to these conditions for rural patients is the ‘early detection of deterioration in symptoms’, which is critical for ‘just in time’ interventions. This planning project is a collaborative effort among the Northern Arizona University (NAU) and University of Michigan (UM) as well as community leaders in rural and tribal health to discuss the best strategies to utilize an integrated remote heart monitoring system to benefit the patients with chronic heart conditions who live in rural, remote and isolated tribal areas.This planning project offers several innovative approaches by working with the tribal AI community to begin(i) Developing a new remote heart monitoring technology to predict the deterioration of the symptoms and occurrence of critical heart conditions in patients with some common chronic cardiac conditions such as atrial fibrillation and congested heart failure. Several remote heart monitoring systems have focused on reliable detection of such events, however by the time that the device alerts the patients or their family, it is already too late to seek medical help for the patients who live in rural and remote areas. Hence, our early prediction framework can give the patients enough time to seek medical assistance. This system can also help caregivers control severe symptoms, reduce readmissions, and reduce the cost associated with care; (ii) Developing deep learning-based and Markov-based prediction methods; and (iii) Developing on-device prediction methods that can work independently of the cloud in order to service the patients with no access to broadband internet. This study can be replicated for a wide range of other diseases and medical conditions and in different geographic regions.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.