Medical information are stored in textual format among the biological data stored in MedLine. With the growing number of medical thesis, research papers, research articles manually extracting this data is a tough work and may lead to some errors. So this work of relational extraction is done by the modern AI and data processing techniques which separate the non-informative content ( i.e. advertisements, scroll bars, menus, citations, quick links, announcements, special credits, etc. ) from the search engines like PubMed.
Thus the process of removing the non-informative content and text mining on the extracted document is discussed further. From the extracted file the information related to the disease is taken and the causes, symptoms and treatments related to the disease are provided to the user. This gives a good quality result and also saves time. AI technology can also be used to decide which kind of treatment to be given to a particular patient when more than one technique are available.