CAREER: Advancing Personal Informatics through Semi-Automated and Collaborative Tracking

这项研究考察了一种名为半自动跟踪的新型自我跟踪方法，以帮助人们轻松地获取丰富的个人数据，如体重、活动、睡眠模式和用药情况。原则上，了解自我跟踪数据可以帮助人们反思自己的健康状况，并了解他们的行为如何影响他们实现目标的进程，从而潜在地改善健康和福祉。然而，在实践中，自我跟踪是困难的。手动跟踪方法，如日记，需要付出很大努力，而自动跟踪方法，如可穿戴式传感工具，与手动跟踪相比，显著降低了跟踪者的意识、责任感和参与度。为了解决这些问题，本研究提出设计和开发一个半自动的跟踪平台，结合手动和自动的数据收集方法。该平台将使人们能够设计和定制自己的跟踪工具，并根据个人不同的跟踪需求捕获多种个人数据。由于定制本身可能很难，该平台还将通过纳入基于专家输入创建的模板来支持协作跟踪。该提案将在与两个可以从自我跟踪中受益的用户群体合作时测试这些想法：(1)生活在退休社区的老年人，以及(2)参与康复计划(即改善患者手术前健康的饮食和锻炼计划)的临床医生和手术患者。老年人和康复小组之间的个体差异、他们的动机和跟踪需求将为如何设计半自动和协作式跟踪工具提供见解。首席调查员(PI)将在她的人机交互和个人信息学课程中使用产生的案例研究和研究平台，并向其他教育工作者和研究人员开放课程。PI还将与她所在机构的项目密切合作，让来自代表性不足群体的人参与研究，包括本科生、女性和少数群体。研究的第一阶段将侧重于通过形成性研究了解人们在使用跟踪技术方面的关切、需求和挑战。研究团队将使用支持可定制手动跟踪的平台，以及对老年人、临床医生和外科患者的采访和观察，进行技术探索性研究。在这一阶段，研究小组将继续开发研究平台，作为拟议研究和干预措施的技术基础。这些形成性研究将使人们深入了解这些人群目前是如何进行自我跟踪的：他们做什么，他们想做什么，什么让他们难以做到，他们害怕什么。这些见解将为研究平台提供一般设计指南，其中将包括半自动跟踪元素，旨在平衡人们的信息需求和数据捕获负担，同时加强他们的参与。在第二阶段，研究小组将通过短期部署研究和设计迭代来测试半自动跟踪方法的可行性。然后，修订后的平台将被纵向部署，以测试其对老年人和外科患者的疗效。在第三阶段，研究团队将专注于协作跟踪方法，旨在帮助患者配置自我跟踪设置，并收集对临床医生有用的高质量数据。为了支持这些目标，研究小组将与临床医生一起举办设计研讨会，以生成常见康复方案的模板，这些模板将在稍后纳入研究平台。该平台的协作跟踪要素将在医院的纵向研究中进行评估。这项研究将有助于不断增长的个人信息学和健康信息学的知识体系。它将向我们介绍老年人和外科患者的自我跟踪实践，以及应如何设计自我跟踪工具来支持包括合作伙伴、照顾者和临床医生在内的各种利益相关者的需求。作为扩大工作影响的一种方式，研究小组将向学术社区(例如行为科学家、个人信息学研究人员)、医疗社区(例如临床医生和患者)、量化自我社区(即专用自我跟踪器)以及希望从事自我跟踪实践的个人传播半自动跟踪平台。

This research examines a novel self-tracking approach called semi-automated tracking to help people easily engage with a rich set of personal data, such as weight, activities, sleep pattern, and medication use. In principle, being aware of self-tracking data can help people reflect on their health condition and understand how their behavior affects their progress toward goals, potentially improving health and well-being. In practice, however, self-tracking is hard. Manual tracking approaches such as diaries require much effort, while automated tracking approaches such as wearable sensing tools significantly reduce the tracker's awareness, accountability, and involvement compared to manual tracking. To address these problems, this research proposes to design and develop a semi-automated tracking platform, combining both manual and automated data collection methods. The platform will enable people to design and customize their own tracking tools and capture many kinds of personal data depending on individuals' diverse tracking needs. As the customization itself can be hard, the platform will also support collaborative tracking by incorporating templates created based on experts' input. The proposal will test these ideas while working with two user groups who can benefit from practicing self-tracking: (1) older adults living in retirement communities, and (2) clinicians and surgical patients in prehabilitation programs (that is, dietary and exercise plans for enhancing patients' health prior to surgery). The differences in individuals, their motivations, and the demands of tracking between the older adult and prehabilitation groups will provide insight on how to design semi-automated and collaborative tracking tools. The principal investigator (PI) will use both the resulting case studies and research platform in her courses on human-computer interaction and personal informatics, and make the curricula openly available to other educators and researchers. The PI will also work closely with programs at her institution to involve people from under-represented groups in the research, including undergraduates, women, and minorities.The first phase of the research will focus on learning people's concerns, needs, and challenges regarding the use of tracking technologies via formative studies. The research team will conduct a technology probing study using a platform that supports customizable manual tracking, along with interviews and observations of older adults, clinicians, and surgical patients. During this phase, the research team will continue developing the research platform, which will be used as the technical basis for the proposed studies and interventions. These formative studies will generate insights into how these populations currently approach self-tracking: what do they do and what would they like to do, and what makes it hard and what are they afraid of. These insights will provide general design guidelines for the research platform, which will include the semi-automated tracking elements designed to balance people's information needs and data capture burden while enhancing their engagement. In the second phase, the research team will test the feasibility of the semi-automated tracking approach with a short-term deployment study followed by design iterations. Then the revised platform will be deployed longitudinally to test its efficacy with both older adults and surgical patients. In the third phase, the research team will focus on the collaborative tracking approach, aiming to help patients configure self-tracking settings and collect high quality data that are useful for clinicians. To support these goals, the research team will conduct design workshops with clinicians to generate templates for common prehabilitation regimens, which will later be incorporated in the research platform. The collaborative tracking elements of the platform will be evaluated in a longitudinal study in hospitals. This research will contribute to the growing bodies of knowledge in personal informatics and health informatics. It will inform us of elders' and surgical patients' self-tracking practices and ways in which self-tracking tools should be designed to support the needs of various stakeholders, including partners, caretakers, and clinicians. As a way to expand the impact of the work, the research team will disseminate the semi-automated tracking platform to academic communities (e.g., behavioral scientists, personal informatics researchers), medical communities (e.g., clinicians and patients), Quantified Self communities (i.e., dedicated self-trackers), and individuals who wish to engage in self-tracking practices.