聊天机器人
对话
构思
医学
以用户为中心的设计
计算机科学
人格
人机交互
万维网
心理学
认知科学
沟通
作者
Mina Jacobs,Ayoposi Ogboye,Nneka Mobisson
标识
DOI:10.1097/01.hjh.0000837428.79504.19
摘要
Objective: To describe how human-centered design was used to create an artificial intelligence (AI) enabled chatbot for people living with hypertension. Design and method: The project focused on technology augmentation and testing to strengthen the mDoc’s CompleteHealthTM platform (an omnichannel platform that provides self-management techniques digitally to people with chronic diseases) to enable us to drive scale in providing self-care support. We leveraged Human-Centered Design principles to guide our pathway based on inspiration, ideation, and implementation to design a chatbot called “Kem’’, aimed to provide real-time response to members living with hypertension. In the inspiration phase, our clinical and tech teams hypothesized that 80% of our growing members’ questions on hypertension could be answered by Kem at scale. In the ideation phase, we leveraged our members to understand and develop the conversation structure and conducted extensive research on how to provide real-time responses to as many members as possible. Kem’s prototype was developed on Botmock (a tool used for designing and fleshing out conversational design flow) with a persona designed to function as a health coach to help bridge the gap in knowledge of health topics, guide, and answer mDoc members. After a series of iterations based on users’ feedback, during the implementation phase, Kem was integrated into a chat platform, Telegram, and was tested with a larger group of users. Results: By applying human-centered design and integrating user feedback into the design and creation of a chatbot to answer questions around hypertension, we were able to create a tool that will help members increase their knowledge of the prevention and management of hypertension. Conclusions: Co-creation should occur at every step of the journey when designing for members’ needs and behaviors. The users’ expressed excitement about Kem and the design process revealed that we needed to be more intentional about making the conversation with the bot interesting, user-friendly, and less technical. As the implementation phase continues and more members use Kem, we will be able to educate them about the prevention and management of hypertension.
科研通智能强力驱动
Strongly Powered by AbleSci AI