Feasibility testing of a novel AI-enabled, cloud-based ECG diagnostic solution to enable fast and affordable diagnosis in long-term continuous ambulatory ECG monitoring

对新型人工智能、基于云的心电图诊断解决方案进行可行性测试，以在长期连续动态心电图监测中实现快速且经济实惠的诊断

• 批准号: 10545691
• 负责人: Bin Fang
• 金额: $ 25.96万
• 依托单位: ZBEATS, INC.
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-12 至 2024-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10545691
• 关键词: Address; Adopted; Affect; Algorithms; Arrhythmia; Award; Businesses; Cardiac; Clinical; Clinical Data; Clinical Research; Cloud Computing; Collaborations; Cost Analysis; Data; Data Analyses; Data Collection; Data Set; Databases; Detection; Devices; Diagnosis; Diagnostic; Disease; Early Diagnosis; Electrocardiogram; Emergency department visit; Enrollment; Environment; Etiology; Evaluation; Event; Future; Goals; Healthcare; Holter Electrocardiography; Hospitalization; Human; Human Resources; Informed Consent; Investigation; Ischemic Stroke; Lead; Life; Medicaid; Medical; Medical center; Medicare; Monitor; Observational Study; Pain; Participant; Patients; Persons; Phase; Phase II Clinical Trials; Physicians; Population; Postdoctoral Fellow; Preventive treatment; Procedures; Protocols documentation; Questionnaires; Recording of previous events; Reporting; Research; Risk; Role; Sample Size; Signal Transduction; Small Business Innovation Research Grant; Small Business Technology Transfer Research; Statistical Data Interpretation; Stream; Stroke; Surveys; Symptoms; Syncope; System; Technology; Telemetry; Telephone; Test Result; Testing; Time; Training; United States National Institutes of Health; Universities; University Hospitals; Work; artificial intelligence algorithm; base; clinical practice; clinical research site; cloud based; commercialization; cost; data exchange; data visualization; design; detection sensitivity; disability; feasibility testing; flexibility; follow-up; high risk; improved; innovation; innovative technologies; mortality; novel; participant enrollment; phase 1 study; phase 2 study; prototype; recruit; remote patient monitoring; risk stratification; safety net; satisfaction; standard of care; stroke patient; success; telehealth; tool; wearable device

PROJECT SUMMARY. The proposed observational study is to evaluate the feasibility of a novel ECG monitoring system leveraging concurrent AI and cloud technologies in long-term continuous monitoring (LTCM) in the clinical environment. It does not intend to use any data or information from the investigational solution to interfere, intervene or affect any clinical decisions made for the participants. Among nearly 2M per year syncope or TIA/stroke patients, 12-15% are cardiac-arrhythmia associated, which usually carries higher risk for long-term disability and even mortality than other-etiologies patients. Proper risk stratification and early initiation of appropriate preventative treatment can result in significant reduction of the cardiac related diseases and their associated mortality. Although LTCM has been proven to be able to detect arrhythmia with high diagnostic yield, the current standard of care has major market pains: 1) days-to-weeks of delay to deliver final report for offline extended Holter; 2) low accuracy in stream arrhythmia detection for online Mobile Cardiac Telemetry; and 3) physicians do not have access to patients’ ECG data. ZBeats’ solution is aiming to improve today’s standard of care by addressing technology accessibility and affordability. ZBPro™, ZBeats’ alpha prototype was validated against our proprietary dataset as well as public datasets required in ANSI/AAMI EC57, demonstrating algorithms, data transmission and visualization work well as expected. In this Phase I study, the feasibility will be tested in the clinical environment by completing the following specific aims (SA): SA1: setup data collection systems and provide training to clinical personnel prior to recruitment. SA2: Conduct patients’ acceptability evaluation by enrolling 60-75 patients to wear the device for up to 7 days. SA3: Evaluate the arrhythmia-capturing capability by conducting physician’s satisfaction questionnaires after reviewing the reports generated from the study system. SA4: Conduct data analysis and start designing the protocol for Phase II study. This proposal will undergo collaboration among ZBeats, Stony Brook University Hospital and Lankenau Medical Center. The long- term goal is to dramatically improve the current standard of care in LTCM by reducing the time to detection of life-threatening arrhythmia from weeks to minutes for cardiac-related high-risk patients, increase the streaming detection accuracy and reducing the total costs by leveraging AI algorithms, cloud infrastructure and a low-cost flexible-material patch. This cost reduction will lead to more general medical use cases, such as telehealth & Remote Patient Monitoring (RPM) to benefit broader population.

项目摘要。拟议的观察性研究是为了评估一种新型心电图监测的可行性 系统利用并发人工智能和云技术进行长期连续监测（LTCM）， 临床环境其不打算使用研究溶液中的任何数据或信息进行干扰， 干预或影响为参与者做出的任何临床决策。在每年近200万例晕厥或 TIA/卒中患者中，12-15%与心律失常相关，这通常具有较高的长期风险 残疾，甚至死亡率高于其他病因的患者。适当的风险分层和早期启动 适当的预防性治疗可以显著减少心脏相关疾病及其 相关死亡率。虽然LTCM已被证明能够以高诊断率检测心律失常， 当前的护理标准具有主要的市场痛苦：1）数天至数周的延迟以交付离线的最终报告 扩展霍尔特; 2）在线移动的心脏遥测的流式心律失常检测准确度低;以及3） 医生无法访问患者的ECG数据。ZBeats的解决方案旨在提高当今的 通过解决技术的可获得性和可负担性来提供护理。ZBPro™，ZBeats的alpha原型得到验证 与我们的专有数据集以及ANSI/AAMI EC 57中要求的公共数据集进行比较，证明 算法、数据传输和可视化工作正常。在第一阶段研究中， 通过完成以下特定目标（SA）在临床环境中进行测试：SA 1：设置数据收集 系统，并在招聘前为临床人员提供培训。SA 2：进行患者的可接受性 通过招募60-75名患者佩戴器械长达7天进行评价。SA 3：评估捕集的捕集效果 通过在审查从生成的报告后进行医生的满意度问卷调查的能力 学习系统SA 4：进行数据分析并开始设计II期研究方案。这项建议会 ZBeats、斯托尼布鲁克大学医院和Lankenau医疗中心进行合作。很长的- 长期目标是通过减少检测时间来显著提高LTCM的当前护理标准。 对于心脏相关高危患者来说，危及生命的心律失常从数周到数分钟不等，增加流媒体 通过利用人工智能算法、云基础设施和低成本的 柔性材料贴片。这种成本降低将导致更多的一般医疗用例，如远程医疗和 远程患者监测（RPM）将使更多人群受益。