Self-Enforcing Networks (SEN) for the development of (medical) diagnosis systems
Christina Klüver · 2016
A Self-Enforcing Network, which is a self-organized neural network, is introduced to develop a differentiated medical diagnosis system. The network is easily trained by data from medical websites and doctors; after training a user can insert the symptoms and obtains a possible diagnosis, or the names of drugs, which cause side effects according to the symptoms. The results show if the symptoms are sufficient to unambiguously identify a specific disease or if there are not enough symptoms to give a safe diagnosis. The described prototype includes concrete examples and shows the potential of such a network for diagnosis systems, in particular for worried unprofessional users.