Clinically relevant machine learning knowledge from bed entry/exit and sleep patterns of older adults.****

来自老年人入床/下床和睡眠模式的临床相关机器学习知识。****

• 批准号: 533698-2018
• 负责人: Bilodeau, Martin
• 金额: $ 1.82万
• 依托单位: University of Ottawa
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=645484
• 关键词: Clinically; relevant; machine; learning; knowledge

Seniors face a number of health-related challenges, including declines in mobility and increased fall risks. Several events can influence how an individual can cope with these challenges. For example, a sudden illness or a poor night sleep can negatively affect mobility. In particular, sleep can become less regular as one ages, and this can lead to mobility that is more difficult periodically. Various sensors have been developed in recent years that allow the monitoring of mobility and sleep patterns in the home setting. At the present time, these sensors can provide targeted health information, but mostly on an individual basis. What is needed to allow an impact at a systems level, where clinically relevant health data can be generated for thousands of Canadian seniors and accessed by healthcare professionals, is a commercial solution that will establish a platform with valid, useful and timely health information.****The partnering organization, IBM, is interested in developing a solution for the monitoring of health-related features of seniors, expanding on their data analytic tools and platforms. Such an application would be of great benefit for Canada's health care system.****The researcher, through a recently completed demonstration project, compiled a database of longitudinal pressure-sensitive mat data from 20 seniors with changing mobility status for approximately 1 year each, in addition to targeted information related to their health. This vast amount of data (big data) represents a unique opportunity to extract health features, particularly mobility- and sleep-related, that would be important and useful for clinicians. The project will look at challenges to create data analytics and classification algorithms that can produce clinically relevant results for clinicians. This would allow optimal review and actions by healthcare professionals without becoming a burden.****We hope that this collaboration will lead to the development of a solution to generate relevant clinical information for seniors and their healthcare team, which would allow IBM to potentially bring such novel solution to market.******

老年人面临着许多与健康相关的挑战，包括行动能力下降和跌倒风险增加。有几个事件可以影响一个人如何科普这些挑战。例如，突发疾病或夜间睡眠不佳可能会对行动能力产生负面影响。特别是，随着年龄的增长，睡眠可能变得不那么规律，这可能导致周期性的流动性更加困难。近年来已经开发了各种传感器，其允许在家庭环境中监测移动性和睡眠模式。目前，这些传感器可以提供有针对性的健康信息，但主要是基于个人的。为了在系统层面产生影响，为数千名加拿大老年人生成临床相关的健康数据，并由医疗保健专业人员访问，需要的是一个商业解决方案，该解决方案将建立一个有效，有用和及时的健康信息平台。合作组织IBM有兴趣开发一种用于监测老年人健康相关特征的解决方案，扩展其数据分析工具和平台。这对加拿大的医疗保健系统有很大的好处 *。研究人员通过最近完成的一个示范项目，编制了一个纵向压敏垫数据数据库，该数据库来自20名老年人，这些老年人的移动状态变化了大约1年，此外还有与他们健康相关的目标信息。大量的数据（大数据）代表了一个独特的机会来提取健康特征，特别是与移动和睡眠相关的特征，这对临床医生来说是重要和有用的。该项目将研究创建数据分析和分类算法的挑战，这些算法可以为临床医生产生临床相关的结果。这将允许医疗保健专业人员进行最佳审查和采取行动，而不会成为负担。我们希望这次合作将导致开发一个解决方案，为老年人及其医疗团队生成相关的临床信息，这将使IBM有可能将这种新颖的解决方案推向市场。