Developing future interfaces for Digital health data

数字医疗保健技术为改善医疗保健服务提供了大量机会。数字医学、人工智能（AI）和机器人技术在医疗保健领域的应用有可能提高诊断、治疗的准确性和护理效率（Topol Review，2019）。数字技术的目标不是取代临床专业人员的专业知识，而是支持和增强他们的活动和决策过程。通过使用这些工具获得的时间应该使工作人员能够花更多的时间与患者在一起并提供更好的护理（Topol Review，2019）。大量的、不必要的或不正确的信息（例如，华而不实的可视化、不正确的比例、不统一的彩色地图）很容易导致认知超载，并对人们解释和比较不同信息来源的能力产生不利影响。增强临床医生决策的数字健康界面应该以这样一种方式设计，以扩展人类的能力并弥补其弱点。这种可视化技术可以与其他技术相结合，如交互式可视化，语音识别，可穿戴技术和预测分析，以促进有效的数据提取。该项目旨在为数字健康开发新的界面，以优化临床医生与数据的交互，重点是如何通过为临床医生提供新的工具和交互方式，使他们更容易获得新兴的健康数据。确定需要和确定要求，以加强目前的做法。2.开发满足这些需求的设计原型，并结合人类认知知识，以扩展功能并弥补弱点。3.通过有效地结合不同的技术来构建设计的迭代版本。4.用户测试，以评估整个过程中产生的工具的有效性。最初，将使用定性方法（访谈，焦点小组），以更好地了解用户的需求和他们面临的当前问题。观察方法将用于早期原型（例如，交互式可视化、语音识别软件、可穿戴技术），以了解如何在上下文中操作技术，确认需求并探索想法。具有机器学习的预测分析也可用于生成个性化治疗算法和风险预测。定量研究方法将在后期阶段证明更有用，并将使用合成患者数据或历史数据进行实验研究，以衡量界面在改善决策过程中的有效性。此外，还将进行用户调查，以根据可用性原则评估接口。

Digital healthcare technologies provide a multitude of opportunities for improvements in healthcare service delivery. The adaptation of digital medicine, artificial intelligence (AI) and robotics into healthcare provision has the potential to improve the accuracy of diagnoses, treatments and the efficiency of care (Topol Review, 2019). The goal of digital technologies is not to replace the expertise of clinical professionals but rather support and augment them in their activities and decision-making processes. The time gained through the use of such tools should enable the staff to spend more time with patients and provide better care (Topol Review, 2019). High-quantity, unnecessary or incorrect information (eg. flashy visualisations, incorrect scaling, non uniform colour maps) can easily lead to cognitive overload and have a detrimental effect on one's ability to interpret and compare different sources of information. Digital health interfaces augmenting clinician's decision making should be designed in such a way to extend human capabilities and compensate for their weaknesses. Such visualisation techniques can be combined with other technologies such as interactive visualisations, speech recognition, wearable technologies and predictive analytics to promote effective data extraction. This project seeks to develop novel interfaces for digital health to optimise the clinician's interactions with data, with a focus on how to make burgeoning health data more accessible to clinicians by providing them with new tools and ways of interacting with it. The following steps are identified to achieve the aims of the project: 1. Identifying needs and establishing requirements to augment current practices. 2. Developing design prototypes that meet these needs and incorporate knowledge of human cognition to extend capabilities and compensate for weaknesses. 3. Building iterative versions of the designs by effectively combining different technologies. 4. User testing to assess the effectiveness of resulting tools throughout the process. Initially, qualitative approaches (interviews, focus groups) to better understand user needs and current issues they are facing will be utilised. Observational methods will be used for early prototypes (eg., interactive visualisations, speech recognition software, wearable technologies) to see how the technology is manipulated in context, confirming requirements and exploring ideas. Predictive analytics with machine learning may also be used to produce personalised treatment algorithms and risk prediction. Quantitative research methods will prove more useful during the later stages and experimental studies using synthetic patient data or historical data will be run to measure the effectiveness of the interfaces in improving the decision making process. In addition, user surveys will be conducted to assess the interfaces in terms of usability principles.