CLINIC-GUIDED COMPUTER DIAGNOSTICS ON THE BASIS OF EXAMPLES

摘要

1. A device for optimizing computer diagnostics based on examples, while the device contains:! a parameter obtaining unit (132) configured to select the optimal set of features of the volume of interest (VOI) so that examples should be invoked so that an objective similarity assessment based on the selected features, clusters in the space of features, volumes of interest (VOI) (116) in the database, consistent with the grouping of the VOIs in the database in the respective clusters through subjective similarity assessment, (S220, S230); and! an output module for providing the selected set as output. ! 2. The device according to claim 1, wherein said selection is that an optimal similarity criterion is selected for use in invoking the examples (S250). ! 3. The device according to claim 2, wherein said choice is that an inter-point distance metric is selected that contains the said optimal similarity criterion, while the said metric is used when grouping in clusters based on an objective assessment, in the said feature space given mentioned selected features (S240). ! 4. The device according to claim 1, in which the signs in said set define the mentioned space of signs (S240). ! 5. The device according to claim 1, wherein said selection comprises iteratively selecting a plurality of VOI features to obtain said optimal set (S230). ! 6. The device according to claim 5, in which the genetic algorithm (400) performs an iterative selection. ! 7. The device according to claim 5, in which this iteration selects the corresponding set that defines the space n