MedLingua: A conceptual framework for a multilingual medical conversational agent

Area of Science:

• Artificial Intelligence in Healthcare
• Natural Language Processing
• Medical Informatics

Background:

• Existing medical chatbots face limitations in handling diverse languages and complex health queries.
• Effective healthcare communication is crucial but often hindered by language barriers, especially in global and rural contexts.

Purpose of the Study:

• To introduce a hybrid AI model for an advanced medical chatbot.
• To enhance the accuracy and promptness of responses to users' health-related queries.
• To overcome limitations of previous medical chatbots, particularly in multilingual support.

Main Methods:

• Development of a hybrid machine learning (ML) model.
• Integration of a dual-stemming approach (P-Stemmer and NLTK-Stemmer) for semitic and non-semitic languages.
• Inclusion of an automatic translation module for multilingual diagnostics.
• Application of the Scrum methodology for agile development.

Main Results:

• The proposed model effectively addresses healthcare communication challenges.
• The dual-stemming approach enhances language processing capabilities for diverse linguistic backgrounds.
• The automatic translation module facilitates a seamless multilingual diagnostic experience.
• The system prioritizes key medical information analysis, including symptoms and prescriptions.

Conclusions:

• The intelligent, multilingual chatbot improves the effectiveness and availability of healthcare services.
• This technology eliminates language barriers, extending healthcare access to underserved regions.
• The hybrid model represents a significant advancement in smart medical assistance and chatbot accuracy.