Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

AsthmaBot: Multi-modal, Multi-Lingual Retrieval Augmented Generation For Asthma Patient Support

Adil Bahaj, Mounir Ghogho

TL;DR

AsthmaBot, a multi-lingual, multi-modal retrieval-augmented generation system for asthma support, is introduced and evaluation of an asthma-related frequently asked questions dataset shows AsthmaBot's efficacy.

Abstract

Asthma rates have risen globally, driven by environmental and lifestyle factors. Access to immediate medical care is limited, particularly in developing countries, necessitating automated support systems. Large Language Models like ChatGPT (Chat Generative Pre-trained Transformer) and Gemini have advanced natural language processing in general and question answering in particular, however, they are prone to producing factually incorrect responses (i.e. hallucinations). Retrieval-augmented generation systems, integrating curated documents, can improve large language models' performance and reduce the incidence of hallucination. We introduce AsthmaBot, a multi-lingual, multi-modal retrieval-augmented generation system for asthma support. Evaluation of an asthma-related frequently asked questions dataset shows AsthmaBot's efficacy. AsthmaBot has an added interactive and intuitive interface that integrates different data modalities (text, images, videos) to make it accessible to the larger public. AsthmaBot is available online via \url{asthmabot.datanets.org}.

AsthmaBot: Multi-modal, Multi-Lingual Retrieval Augmented Generation For Asthma Patient Support

TL;DR

AsthmaBot, a multi-lingual, multi-modal retrieval-augmented generation system for asthma support, is introduced and evaluation of an asthma-related frequently asked questions dataset shows AsthmaBot's efficacy.

Abstract

Asthma rates have risen globally, driven by environmental and lifestyle factors. Access to immediate medical care is limited, particularly in developing countries, necessitating automated support systems. Large Language Models like ChatGPT (Chat Generative Pre-trained Transformer) and Gemini have advanced natural language processing in general and question answering in particular, however, they are prone to producing factually incorrect responses (i.e. hallucinations). Retrieval-augmented generation systems, integrating curated documents, can improve large language models' performance and reduce the incidence of hallucination. We introduce AsthmaBot, a multi-lingual, multi-modal retrieval-augmented generation system for asthma support. Evaluation of an asthma-related frequently asked questions dataset shows AsthmaBot's efficacy. AsthmaBot has an added interactive and intuitive interface that integrates different data modalities (text, images, videos) to make it accessible to the larger public. AsthmaBot is available online via \url{asthmabot.datanets.org}.
Paper Structure (17 sections, 2 figures, 6 tables)

This paper contains 17 sections, 2 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Previous Literature
  3. Large Language Models (LLMs)
  4. Biomedical LLMs and RAG
  5. Methodology
  6. Data Collections
  7. Vector DB Building and Retrievers
  8. Prompting and Backbone LLM
  9. Translation
  10. Visual Interface
  11. AsthmaBot's Inference Process
  12. Results
  13. Evaluation Procedure
  14. Evaluation Data
  15. Evaluation Results
  16. ...and 2 more sections

Figures (2)

  • Figure 1: AsthmaBot interface.
  • Figure 2: AsthmaBot Overview: given a query AsthmaBot uses asthma-related external resources to retrieve information relevant to the query and the query language before synthesizing a multi-modal response. This framework provides a multi-modal response grounded in truth, which reduces hallucinations and provides multiple formats for the answer.