A Disease Prediction and Medication Recommendation System Using Machine Learning Techniques

The fusion of technology and healthcare has heralded a new era of patient-centric solutions. This work presents a comprehensive platform that predicts infectious and systemic diseases such as viral infections, autoimmune disorders, and gastrointestinal conditions using three machine learning algorithms namely Naive Bayes, Random Forest and Decision Tree. In addition to disease prediction, the platform recommends appropriate drugs and facilitates seamless appointment bookings with healthcare professionals. The system is trained with datasets focused on these specific types of diseases, making it highly effective within this context. Enhancing this digital experience is a call bot, enabling users to interact via offline phone call speech-to-text-to-speech functionality. This addition ensures greater accessibility, catering to a broader user demographic, especially those less familiar with traditional interfaces. At its core, the Naive Bayes algorithm, renowned for its efficiency and effectiveness, was employed. Upon rigorous data preprocessing, the model was trained to offer predictions based on patient symptoms and history. Complementing this, the platform can recommend appropriate drugs for identified ailments. Beyond predictions and recommendations, the platform offers an integrated appointment booking system. Users can effortlessly schedule, reschedule, or cancel appointments with doctors. To bolster user adherence to these appointments, a reminder system is in place, which proactively sends notifications to users through their preferred communication channel, be it email or text message. Preliminary user interactions indicate commendable accuracy in disease prediction and drug recommendation. The call bot effectively captures and translates user commands, while the reminder system has been proven to improve appointment adherence significantly.