In the realm of healthcare, the accurate prediction and diagnosis of diseases play a pivotal role in enhancing patient outcomes and optimizing resource allocation. This paper explores the utilization of Large Language Models (LLMs) as a transformative tool in disease prediction and diagnosis. Leveraging patient data, medical records, and symptoms, we investigate the efficacy of LLMs in analysing unstructured textual information to infer potential diseases. Our methodology involves preprocessing the data, fine-tuning pre-trained LLMs, and evaluating their performance using standard metrics. Through experimentation, we demonstrate the effectiveness of LLMs in predicting and diagnosing diseases with high accuracy. Furthermore, we explore the extensive applications of LLMs within healthcare, emphasizing their capacity to transform decision-making workflows and enhance patient care quality. Overall, this research sheds light on the promising role of LLMs in advancing disease prediction, diagnosis, and ultimately, healthcare transformation.