Method For Processing A Radiological Image

Abstract: Radiological-image-processing method for processing a radiological image /(x, y), in the digital format, comprising at least one radiological anomaly detected using a convolutional neural network trained to detect radiological anomalies in radiological examinations, the radiological image /(x, y) being characterised by the brightness of each of its pixels /(x,y), and by at least one radiological-anomaly-influence map Ck(x,y) that attributes, to each pixel (x,y) of the radiological image /(x,y), a value representative of the proportion in which the pixel had an influence on the detection result of the radiological anomaly k, which method is computer-implemented, and comprises steps of: - normalising (3) the radiological-anomaly-influence maps Ck(x,y) so as to obtain normalised radiological-anomaly-influence maps Ckn(x,y); - fusing (4) the normalised radiological-anomaly-influence maps Ckn(x,y) into a single fused influence map C(x,y); - carrying out processing (5) to improve the image /(x, y), this exploiting a brightness histogram, wherein the contribution of each pixel to the computation of the brightness histograms is weighted by the fused influence map C(x,y).

Patent Information

Applicants

Inventors

Specification

Documents

Application Documents

Search Strategy