Applying text mining and natural language processing methods for medical text documents are a modern and very promising approach. It has been quite a long time since hospitals started using digital ways for written text document creation instead of handwriting, so they have produced a huge amount of domain specific data. However the meaning of digitalisation has totally changed for today. It is a reasonable expectation to be able to search and find trustworthy information, possibly related knowledge and so on.
Another aspect of processing medical documents is that instead of textbooks and scientific papers, these records describe cases of real people in real life. As such it is reasonable to suspect that there are many hidden relations, patterns, complex events and even standalone information hidden from the surface reader that could and should be extracted. In modern medicine the use of multimodal information has also risen deploying the need of processing images, sound and video streams.
There have been some attempts aiming at such a goal, but a fully comprehensive system, especially in a multilingual environment is a unique initiative. Our system will handle special linguistic phenomena by applying advanced text processing -- in the case of written documents -- and will use several kinds of expert knowledge provided by both human and machine resources for all types of multimodal information processing as well. By being built on such a base, it will be able to apply deeper semantic analysis and finally present a multifunctional usage, that is evaluated by requirements that are set up by domain experts.