Data Science Applications in Communication andSwallowing Disorders

项目概要/摘要 电子医疗记录、大型数据注册和易于访问的受保护服务器的出现， 导致数字信息爆炸，对改善患者的临床影响可能很大， 管理和成果。大数据仓库，用于捕获范围内的标准化信息 临床实践使受过训练的科学家不仅可以从事传统的假设检验，而且还可以 发现新的假设，完善现有的理论，并将新的发现应用于健康评估， 干预措施。尽管这些数据平台的可访问性和潜在影响，临床科学家 传统的定向实验包括相对较小的样本量和来自个体的数据， 实验室，并且没有接受过大数据分析方面的培训，也没有参与过适当的团队科学家， 在这个领域工作，如计算机科学家，生物统计学家和工程师。这个项目的首要目标是 建议是指导早期以患者为导向的沟通和吞咽科学家进行大数据分析， 指导并让早期的数据科学学者参与沟通和吞咽研究。PI建议 四个主要的指导和研究目标，在这个K24更新建议：1。培养早期干部 通信和吞咽科学家在数据科学方法，包括机器学习，由专家， 跨学科、协作的数据科学团队，2.参与并引入来自以下领域的早期职业数据科学家 从生物统计学、计算机科学和工程到通信和吞咽科学， 各自的数据集，以促进跨学科的数据科学团队和研究生产力，3.适用 新的数据科学方法，以确定吞咽障碍的表型和严重程度分类， 已知处于与吞咽困难相关的营养和健康并发症的高风险的患者组，以及4. 开发机器学习应用的新研究领域，以提高生理学的可靠性。 吞咽评估。职业发展和研究计划的数据科学主题直接与 NIDCD的数据科学战略计划列出了其使命：存储，管理，标准化和 发表大量的生物医学研究数据。NIDCD认识到， 组织良好、安全和有效运行的数据资源对现代科学研究至关重要， 通过最大化NIH资助的工作产生的数据的价值，生物医学发现的速度 并且可以加速医学突破以获得更好的健康结果。

PROJECT SUMMARY/ABSTRACT The emergence of electronic medical records, large data registries and readily accessible, protected servers have resulted in an explosion of digital information with potentially high clinical impact for improving patient management and outcomes. Big data warehouses that capture standardized information within the scope of clinical practices allow trained scientists to not only engage in traditional hypothesis testing, but to also uncover new hypotheses, refine existing theories and apply new discoveries to health assessments and interventions. Despite the accessibility and potential impact of these data platforms, clinician scientists have traditionally directed experiments that incorporate relatively small sample sizes and data from individual laboratories, and have not been trained in big data analytics or in engaging appropriate team scientists who work in this space, such as computer scientists, biostatisticians and engineers. The overarching goal of this proposal is to mentor early patient oriented communication and swallowing scientists in big data analytics and to mentor and involve early data science scholars in communication and swallowing research. The PI proposes four primary mentorship and research goals in this K24 renewal proposal: 1. Train a cadre of early stage communication and swallowing scientists in data science methods, including machine learning, by an expert, interdisciplinary, collaborative data science team, 2. Engage and introduce early career data scientists from fields of biostatistics, computer science and engineering to communication and swallowing sciences, and respective data sets, toward facilitating interdisciplinary data science teams and research productivity, 3. Apply novel data science methods to identify phenotypes of swallowing impairment and severity classifications in patient groups known to be at high risk for nutritional and health complications related to dysphagia, and 4. Develop a new area of research in machine learning applications toward improving reliability of physiologic swallowing assessment. The data science theme of the career development and research plan directly align with NIDCD's Strategic Plan for Data Science which lists as its mission: Storing, managing, standardizing and publishing the vast amounts of data produced by biomedical research. NIDCD recognizes that accessible, well-organized, secure and efficiently operated data resources are critical to modern scientific inquiry…and by maximizing the value of data generated through NIH-funded efforts, the pace of biomedical discoveries and medical breakthroughs for better health outcomes can be accelerated.