Advanced digital pathology and provisions for remote diagnostics

摘要

The present invention discloses a method for detecting and identifying pathologies in a magnified captured image. The method comprising the step of: performing macro image analysis for identifying abnormal and normal segments of the captured image, performing conversion of colored images to gray scale image, performing segmentation of the gray scale colored biopsy by applying two segmentation levels, merging the image results of the coarse level and the fine level segmentations by expanding the coarse image to fit scale of the fine image and identifying pixels having the same value at both levels, performing comparison between object's properties and characteristics appearing in abnormal segments and object's properties and characteristics appearing in normal segments, and calculating the deviations of each property between the abnormal segments and the normal segments and ranking objects based on the calculated deviations of each property and characteristic.