Design, development and usability of an educational AI chatbot for People with Haemophilia in Senegal.

ABSTRACT

INTRODUCTION: Gaps in the disease knowledge of People with Haemophilia (PWH) in Senegal are important barriers to the effective management of haemophilia. Digital health systems for chronic diseases in low- and middle-income countries are suggested to improve education and self-management. Artificial Intelligence (AI) chatbots could improve knowledge and support symptom monitoring. AIM: Development process and usability testing of an AI chatbot to assess its future adoption in Senegal. METHODS: An AI chatbot prototype was designed based on a multilingual conversational engine using Natural Language Processing. A sequential mixed method was used including a co-creative design process with a task force made up of PWH and medical doctors. Usability was assessed through the System Usability Scale (SUS) questionnaire. RESULTS: An AI chatbot in French and Wolof, named Saytù Hemophilie, was developed for Android and Apple iOS devices. It was assessed as a very usable system with a SUS score of 81.7, above average. 42% would prefer to use the Wolof version even if they were very satisfied with the French version. The level of Wolof in the app did not always correspond to users' levels. Participants praised its accessibility and reliability, and its ability to enhance self-learning. CONCLUSIONS: Findings suggest that a culturally adapted digital conversational agent is likely to be used by PWH in Senegal and their families to improve education and self-management of haemophilia. Relevance and impact are foreseen for other communities in Africa and beyond.