AI-Chatbot for Disease Prediction

Abstract This survey explores the role of AI-powered medical chatbots in disease prediction, with a focus on addressing the limitations of earlier models based on Long Short-Term Memory (LSTM) networks. These earlier systems often suffer from restricted disease coverage and suboptimal prediction accuracy. To overcome these issues, the use of Recurrent Neural Networks (RNNs) is proposed, as they offer improved handling of sequential patient data and the ability to generate more precise health-related responses. The review also highlights the importance of integrating comprehensive disease datasets and advanced natural language processing (NLP) techniques to enrich patient interactions. However, current implementations face persistent challenges, including language variability and inconsistent responses. The study further examines emerging approaches such as hybrid AI systems and modular chatbot architectures, which promise greater flexibility and scalability. Key challenges identified include limited data availability, scalability constraints, and the need for higher diagnostic accuracy. The findings suggest that adopting RNN-based frameworks and training with diverse medical datasets can significantly enhance the effectiveness of medical chatbots, thereby strengthening healthcare delivery and patient engagement.. Keywords— AI, LSTMs, RNNs, NLP