Trustworthy Medical Sentiment Detection for Maternal and Neonatal Healthcare

Abstract

The increasing availability of online medical platforms has made it easier for people to access medical information and treatment. However, it has also led to an increase in fraudulent schemes that exploit individuals seeking medical advice. This has created an exploitable opportunity for unprofessional and unqualified medical personnel operating on online platforms for telemedicine. This study illustrates and discusses the use of Artificial Intelligence, specifically Natural Language Processing (NLP), to detect trustworthy medical sentiments in online maternal and neonatal healthcare advice. Interpretable detection of medical sentiments in crowdsourced advice from both medical experts and regular individuals on social media platforms was done. In this approach, the “Explain Like I'm 5” (ELi5) technique is used to make the detection process more understandable and trustworthy. Our findings demonstrate an urgent need for a maternal and neonatal medical corpus and the use of explainable AI to ensure a sustainable and trustworthy healthcare for all with Conversational AI.