A method and apparatus for performing fluoroscopic noise reduction

摘要

The present invention provides an adaptive fluoroscopic noise reduction (FNR) algorithm for performing fluoroscopic noise reduction on image data acquired by an x-ray fluoroscopy system. The FNR algorithm utilizes various data acquisition parameters to generate an estimation of the noise statistics associated with the fluoroscopic x-ray detector component of the x-ray fluoroscopy system. The FNR algorithm utilizes the data acquisition parameters and knowledge about an object to be used in an x-ray fluoroscopy procedure to predict a minimal object contrast associated with the object. The estimation of the noise statistics and the prediction of the object contrast are then utilized to adapt certain variables that are utilized by the FNR algorithm. The estimation of noise statistics and the prediction of object contrast may be adapted on a pixel-by-pixel basis by using a regional mean to obtain an adapted estimation of noise statistics and an adapted prediction of object absolute contrast. The variables of the FNR algorithm are then adapted based on the adapted noise statistics estimation and the adapted prediction of object absolute contrast.