Bridge2AI: Voice as a Biomarker of Health - Building an ethically sourced, bioaccoustic database to understand disease like never before

Bridge2AI：声音作为健康的生物标志物 - 建立一个符合道德规范的生物声学数据库，以前所未有的方式了解疾病

• 批准号: 10473236
• 负责人: Yael Emilie Bensoussan
• 金额: $ 381.73万
• 依托单位: UNIVERSITY OF SOUTH FLORIDA
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-01 至 2026-08-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10473236
• 关键词: Acoustics; Address; Adoption; Alzheimer' s Disease; Amplifiers; Apple; Attention; Benign; Biological Markers; Bipolar Disorder; Categories; Childhood; Chronic Obstructive Pulmonary Disease; Clinical; Cloud Computing; Collaborations; Communities; Competence; Computer software; Consent; Data; Data Analyses; Data Collection; Databases; Development; Diagnosis; Disease; Educational Curriculum; Electronic Health Record; Engineering; Ensure; Ethics; FAIR principles; Fostering; Future Generations; Generations; Genomics; Guidelines; Health; Heart failure; Human; Infrastructure; Larynx; Lesion; Link; Literature; Malignant neoplasm of larynx; Medical; Mental Depression; Mental disorders; Mentorship; Modeling; Mood Disorders; Neurodegenerative Disorders; Outcome; Paralysed; Parkinson Disease; Pathology; Patients; Pneumonia; Population; Research; Research Personnel; Respiration Disorders; Schizophrenia; Scholarship; Source; Speech Delay; Standardization; Stroke; Technology; Validation; Voice; Voice Quality; Workforce Development; artificial intelligence algorithm; autism spectrum disorder; base; clinical application; clinical care; data acquisition; data preservation; data privacy; data sharing; federated learning; innovation; insight; multidisciplinary; multimodality; nervous system disorder; patient privacy; predictive modeling; privacy protection; public health relevance; radiomics; research and development; screening; skill acquisition; smartphone Application; software infrastructure; tool; tool development; trustworthiness; underserved community; user-friendly; vocal cord

Our group aims to integrate the use of voice as biomarker of health in clinical care by generating a substantial multi-institutional, ethically sourced, and diverse voice database linked to multimodal health biomarkers to fuel voice AI research and build predictive models to assist in screening, diagnosis, and treatment of a broad range of diseases. Data collection will be made possible by software through a smartphone application linked to electronic health records (EHR) and other health biomarkers such as radiomics, and genomics, and supported by federated learning technology to protect data privacy. Based on the existing literature and ongoing research in different fields of voice research, our group has identified 5 disease categories for which voice changes have been associated to specific diseases and around which we aim to center the data acquisition efforts: 1. Vocal Pathologies (Laryngeal cancers, Vocal fold paralysis, Benign laryngeal lesions) 2. Neurological and Neurodegenerative Disorders (Alzheimer’s, Parkinson’s, Stroke, ALS) 3. Mood and Psychiatric Disorders (Depression, Schizophrenia, Bipolar Disorders) 4. Respiratory disorders (Pneumonia, COPD, Heart Failure, OSA) 5. Pediatric diseases (Autism, Speech Delay) Specific Aim #1: Data Acquisition Module: - To build a multi-modal, multi-institutional, large scale, diverse and ethically sourced human voice database linked to other biomarkers of health that is AI/ML friendly to fuel voice AI research Specific Aim #2: Standard Module: - To introduce the field of acoustic biomarkers by developing new standards of acoustic and voice data collection and analysis for voice AI research. Specific Aim #3: Tool Development and optimization - To develop a software and cloud infrastructure for automated voice data collection through a smartphone application that allows non-invasive, user-friendly, high quality voice data collection while minimizing human manipulation. This will include integrated acoustic amplifiers and acoustic quality standardization. - To implement Federated Learning technology to allow analysis of multi-institutional data while minimizing data sharing and preserving patient privacy Specific Aim #4: Ethics Module - To integrate existing scholarship, tools, and guidance with development of new standard and normative insights for identifying, anticipating, addressing, and providing guidance on ethical and trustworthy issues from voice data generation and AI/ML research and development to clinical adoption and downstream health decisions and outcomes. - To develop new guidelines for consenting to voice data collection, voice data sharing and utilization in the context of voice AI technology Specific Aim # 5: Teaming Module: - To build bridges between the medical voice research world, the acoustic engineers, and the AI/ML world to promote the integration of tangible clinical application for Voice AI algorithms Specific Aim #6: Skills and Workforce Development Module - To develop a unique curriculum on voice biomarkers of health and the development, validation, and implementation for AI models that are FAIR and CARE - To create a community of voice AI researchers, especially those from underserved communities, and foster collaborations to promote application of ML for Voice Research - To engage a broad range of learners with competency assessment and mentorship

我们小组的目标是通过生成与多模式健康生物标记相关的大量多机构、道德来源和多样化的语音数据库，将语音作为健康生物标记的使用整合到临床护理中，以推动语音人工智能研究并建立预测模型，以协助筛查、诊断和治疗多种疾病。数据收集将通过与电子健康记录（EHR）和其他健康生物标记（例如放射组学和基因组学）链接的智能手机应用程序的软件来实现，并得到联邦学习技术的支持以保护数据隐私。 根据现有文献和不同语音研究领域正在进行的研究，我们小组确定了 5 种疾病类别，其中语音变化与特定疾病相关，我们的目标是围绕这些疾病类别开展数据采集工作： 1. 声乐病理学（喉癌、声带麻痹、良性喉部病变） 2. 神经系统和神经退行性疾病（阿尔茨海默氏症、帕金森氏症、中风、ALS） 3.情绪和精神疾病（抑郁症、精神分裂症、双相情感障碍） 4. 呼吸系统疾病（肺炎、慢性阻塞性肺病、心力衰竭、阻塞性睡眠呼吸暂停） 5.儿科疾病（自闭症、言语迟缓） 具体目标#1：数据采集模块： - 建立一个多模式、多机构、大规模、多样化和道德来源的人类声音数据库，与其他健康生物标志物相链接，该数据库对人工智能/机器学习友好，以推动语音人工智能研究 具体目标#2：标准模块： - 通过为语音人工智能研究制定声学和语音数据收集和分析的新标准，引入声学生物标记领域。 具体目标#3：工具开发和优化 - 开发一个软件和云基础设施，通过智能手机应用程序自动收集语音数据，从而实现非侵入性、用户友好、高质量的语音数据收集，同时最大限度地减少人为操作。这将包括集成声学放大器和声学质量标准化。 - 实施联邦学习技术，以允许分析多机构数据，同时最大限度地减少数据共享并保护患者隐私 具体目标#4：道德模块 - 将现有的学术、工具和指导与新标准和规范见解的开发相结合，以识别、预测、解决和提供有关道德和可信问题的指导，从语音数据生成和人工智能/机器学习研究和开发到临床采用和下游健康决策和结果。 - 制定语音人工智能技术背景下同意语音数据收集、语音数据共享和使用的新指南 具体目标#5：团队模块： - 在医学语音研究领域、声学工程师和人工智能/机器学习领域之间架起桥梁，促进语音人工智能算法的实际临床应用的整合 具体目标#6：技能和劳动力发展模块 - 开发关于健康语音生物标志物的独特课程，以及公平和关怀的人工智能模型的开发、验证和实施 - 创建一个语音人工智能研究人员社区，特别是来自服务不足社区的研究人员，并促进合作以促进机器学习在语音研究中的应用 - 通过能力评估和指导吸引广泛的学习者