AbstractAccess to proper healthcare services is still difficult in many rural and semi-urban areas, especially for people who are not comfortable using English. Most digital healthcare systems depend on English text input and require basic digital skills, which limits their usability in large populations. To overcome this limitation, a Smart Bilingual Medical Assistant for Disease Detection and Emergency Management was developed. The system supports both Kannada and English and allows users to interact via text, voice, and video. The assistant uses Artificial Intelligence, Natural Language Processing, and Machine Learning techniques to understand symptoms, predict diseases, and provide basic medical guidance. Whisper is used to convert voice input into text, whereas the Google Translate API enables Kannada-to-English translation. Disease prediction was performed using Logistic Regression and Random Forest models. The system also includes emergency alert detection and real-time doctor consultations using Jitsi Meet. Testing with local users showed that the system works effectively and can improve healthcare accessibility in regional areas.
Wali et al. (2026) studied this question.