Development of an automated patient identification method based on an image-matching technique of chest radiographs in PACS environment

摘要

For maintaining its reliability of the PACS, wrong patient's images should not be stored in the improper patients' folders. Our aim is to develop an automated warning system based on an image-matching technique for comparison with previous chest radiographs, in correcting wrong chest radiographs being stored in the PACS. For identification of a wrong patient, the correlation value is determined for a previous image of a patient and a new current image of the presumed corresponding patient. Our preliminary results indicated that more than 50 % of wrong images can be identified correctly with our method. We believe that this automated warning system for patient identification would be useful in correcting wrong images being stored in the PACS environment, and can avoid reading of wrong patient's images during interpretation.