In the modern era, retrieving healthcare information has become a significant global concern. The identification of biological named entities (NER), which has a variety of applications, is a critical step in text mining of medical information. Deep learning-based ways to tackle this activity have been greatly growing and extending recently because its settings can be entirely academic from start to finish without the requirement for manually created characteristics. The Named Entity Recognition method is used in this study to find disease related synonyms to mine the meaning in medical reports and other applications. The Named Entity Recognition algorithm is one of the automated methods for obtaining medical data from an ontology web search. The desired format is prepossessed and changed to the healthcare text input data after it has been downloaded from the Kaggle repository. After pre-processing, the medical data is recovered using the current clustering techniques Particle Swarm Optimisation (PSO) and Fuzzy C Means (FCM), and the results are evaluated. The research effort also proposes PSFCM, a novel technique for extracting medical data. The results of PSFCM are compared to those of PSO and FCM, which are currently in use. Precision, recall, and f-measure values are a few examples of classification methods that are used to evaluate the PSFCM's performance.