CASE BASED REASONING VERSUS ARTIFICIAL NEURAL NETWORKS IN MEDICAL DIAGNOSIS

摘要

Embedding Machine Learning technology into Intelligent Diagnosis Systems adds a new potential to such systems and in particular to the imagiology ones. In our work, this is achieved using the data acquired from MEDsys, a computational environment that supports medical diagnosis systems that use an amalgam of knowledge discovery and data mining techniques, which use the potential of an extension to the language of Logic Programming, with the functionalities of a connectionist approach to problem solving using Artificial Neural Networks. One's goal aims to conceive an alternative method to detect medical pathologies, as an alternative to the one in use in the actual medical diagnostic system; i.e., Case Based Reasoning versus Artificial Neural Networks. A comparative study of these two approaches to machine learning will be presented, taking into account its applicability in MEDsys.