Hierarchical self-learning system for computerized clinical diagnosis support

摘要

Individual computer diagnostic support (CDS) systems are coupled to a “global” CDS system. Each CDS system uses the same learning system or the same learning system technology. Training cases and test cases from each CDS system are provided to the global CDS system, which uses the training cases to generate learning system parameters based on the training cases. By having more training cases of any CDS system than any CDS system, the parameters provided by the global CDS system provide a higher quality diagnostic output than any CDS system of the individual CDS system. The learning system parameters in the global CDS system are provided to each CDS system to update the learning system parameters of the individual CDS system. The global CDS system can also improve and / or adjust the structure of the embodied learning system.