Page 94 - AI for Good Innovate for Impact
P. 94
AI for Good Innovate for Impact
Use Case 10: AI-mediated Interactive Health Messaging for
Community Health Promotion in Low-Middle Income Countries�
Country: Lebanon
Organization: American University of Beirut
Contact Person(s):
Imad Elhajj – ie05@ aub .edu .lb
Aline Germani – ag24@ aub .edu .lb
Michelle El Kawak – mk290@ aub .edu .lb
1 Use Case Summary Table
Item Details
Category Healthcare,
Problem Addressed Residents in low- and middle-income countries (LMICs) such as
Lebanon often face inequitable access to crucial health informa-
tion. This disparity is driven by a confluence of factors, including
language barriers, low literacy rates, pervasive stigma, and ongo-
ing societal crises. As a result, individuals are often unable to make
informed health decisions, leading to poorer health outcomes and
exacerbating existing health disparities [1][2].
Key Aspects of Solution Multilingual chatbot; Speech-to-Text (STT) and Text-to-Speech
(TTS) integration; WhatsApp-based deployment; Retrieval
Augmented Generation (RAG) with Reasoning + Action (ReAct
prompting).
Technology Keywords Large Language Model (LLM), STT, TTS, ReAct prompting, Lang-
Chain, Retrieval Augmented Generation (RAG), WhatsApp API,
vector databases
Data Availability Publicly sourced health content—World Health Organization
(WHO), International Federation of Red Cross and Red Crescent
Societies (IFRC), United Nations Children's Fund (UNICEF); Private
de-identified interaction logs
Metadata (Type of Data) Text and Audio
58
