Deep learning for algorithmic detection of pulmonary hypertension using a combined digital stethoscope and single-lead electrocardiogram

使用组合数字听诊器和单导联心电图进行肺动脉高压算法检测的深度学习

• 批准号: 10547726
• 负责人: Gaurav Choudhary
• 金额: $ 78.7万
• 依托单位: EKO DEVICES, INC.
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-15 至 2025-08-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10547726
• 关键词: Address; Adult; Algorithms; Award; Cardiac; Cardiac Catheterization Procedures; Caring; Cellular Phone; Chest; Clinical; Complex; Data; Data Set; Databases; Detection; Diagnosis; Diagnostic; Disease; Early Diagnosis; Echocardiography; Eko; Electrocardiogram; Enrollment; Evaluation; Functional disorder; Goals; Gold; Health Services Accessibility; Hospitalization; Human Resources; Investigation; Label; Lead; Location; Marketing; Measures; Medical Device; Modality; Morbidity - disease rate; Outcome Study; Patient-Focused Outcomes; Patients; Performance; Personal Computers; Phase; Phenotype; Prognosis; Public Health; Pulmonary Hypertension; Quality-Adjusted Life Years; Randomized; Running; Screening procedure; Sensitivity and Specificity; Severities; Signal Transduction; Small Business Innovation Research Grant; Specificity; Stethoscopes; Stream; Symptoms; Tablet Computer; Testing; Time; Training; Validation; base; care costs; clinical decision support; cloud software; comorbidity; cost; deep learning; deep learning algorithm; design; digital; improved; innovation; mortality; novel; patient screening; point of care; pressure; prototype; pulmonary arterial pressure; research clinical testing; screening; support tools; trial comparing

Abstract This SBIR Direct to Phase II project will develop a deep learning-based clinical decision support algorithm for screening and detecting pulmonary hypertension (PH) using phonocardiogram and electrocardiogram data recording using the Eko DUO Digital Stethoscopes. The screening tool will help to decrease the number of patients with pulmonary hypertension that remain undertreated simply because their condition is not diagnosed. The gold standard diagnostic for pulmonary hypertension is right heart catheterization which is costly, invasive, and requires specialized personnel.To address these challenges, Eko developed the DUO, a digital stethoscope in a handheld form factor with built-in single lead electrocardiogram. The DUO is designed to stream digitized phonocardiograms and electrocardiograms to a smartphone, tablet or personal computer. There, the signal can be analyzed with the decision support algorithm we developed as part of this project. The specific aims of this study are : (1) build a database of matched ECG/PCG recording labeled against right heart catheterization and echocardiograms, and (2) develop and clinically test a deep learning algorithm that can detect PH and stratify its severity. By integrating this deep learning algorithm into Eko’s mobile and cloud software platform, we anticipate this algorithm will enable more accurate screening for pulmonary hypertension in adult patients, leading to earlier diagnosis and better patient outcomes.

摘要 这个SBIR Direct to阶段II项目将开发基于深度学习的临床决策支持 利用心音图和超声心动图筛选和检测肺动脉高压的算法 使用Eko Duo数字听诊器记录心电数据。筛选工具 将有助于减少仍然患有肺动脉高压的患者的数量 仅仅因为他们的病情没有得到诊断，就没有得到足够的治疗。诊断的黄金标准 肺动脉高压是一种费用昂贵、有创性的右心导管术，需要 专业人员。为了应对这些挑战，Eko开发了这两个数字 听诊器为手持式，内置单导联心电图机。二人组是 旨在将数字化心音和心电图流传输到智能手机、平板电脑 或个人计算机。在那里，可以使用决策支持算法WE来分析信号 作为该项目的一部分开发。这项研究的具体目的是：(1)建立数据库 与右心导管术和超声心动图相对照的匹配的CG/PCG记录， (2)开发并临床测试了一种能够检测PH并对其进行分层的深度学习算法 严肃性。通过将这种深度学习算法集成到EKO的移动和云软件中 平台上，我们预计该算法将使肺的筛查更加准确 成年患者的高血压，导致更早的诊断和更好的患者结果。