African Health Stories
Leveraging generative AI to create interactive visual, oral and text stories to guide people living with Type II Diabetes on lifestyle changes
The African Health Stories project leverages generative AI to help people with Type II Diabetes make personalized lifestyle changes to improve their health outcomes. Diabetes is a significant global health challenge, affecting 10% of the population. In South Africa, 12.7% of people aged 20 to 79 were diagnosed with diabetes in 2019, with more than half having Type 2 diabetes. This increase is driven by socio-economic, demographic, environmental, and genetic factors. Research shows that lifestyle changes can halve the risk of Type 2 diabetes and improve health outcomes where prevention isn’t possible. However, current lifestyle advice is often generic and not suited to all patients, making it difficult to apply. Additionally, in lower and middle income countries communication barriers and limited clinic time reduce the effectiveness of provider-patient communication.
Through this project, we aim to use generative AI to create culturally and linguistically appropriate health stories starting with patients in the Western Cape of South Africa. Patients and providers will be able to generate different health stories throughout the diabetes journey, to address different lifestyle aspects. Patients will also be able to interact with the generated stories so that the advice can be tailored to their needs and preferences. In this way, we aim to help patients better understand how to apply care advice to their lives.
We’re collaborating with Stellenbosch University, (opens in new tab) University of Pretoria (opens in new tab), and Swansea University (opens in new tab) to create a story generation tool which produces interactive visual, oral, and text stories enabling patients to better understand how to apply health advice to their local circumstances. We are shortly going to begin testing out stories addressing key diet and exercise questions with 300 participants in the Western Cape of South Africa. As part of this process, we will be working with a team of medical experts as we develop and verify the accuracy of automated checkers.