Clinical pictures of unknown origin in neurology: past, present and future usefulness of artificial intelligence.

Although, in the course of the last 50 years, the achievements in the medical field have been astonishing, at the beginning of the third millennium a number of clinical pictures are still left without a precise nosographic origin. In the past, the delay in scientific communication was the main explanation presented for the lack of understanding of clinical pictures of unknown nosographic origin. The history of medicine provides excellent examples of this dispersion of human capital, even if the history of clinical neurology presents "exceptions" (the pictures that we now call de la Tourette's syndrome and Parkinson's disease) that indicate that major clinical syndromes could be clearly detected and relatively rapidly diffused even in the 19th century. Contrary to the past, the delay in scientific communication no longer seems an obstacle to the sharing of medical knowledge. Nevertheless, the problem of the in-depth comprehension of clinical pictures of unknown nosographic origin still remains dominant, mainly because of the limited spread of ample and flexible online accessible databases of unknown nosographic origin clinical syndromes. The need for interactive electronic archives and other artificial intelligence resources in order to promote progress in clinical knowledge is discussed in this paper.