AUTOMATED, MICROSCOPE-ASSISTED EXAMINATION PROCESS OF TISSUE OR BODILY FLUID SAMPLES

Method for the automated, microscope-aided examination of tissue samples or samples of body fluids with the aid of neural networks. In a first method of examination the sample is firstly classified according to its type and subsequently a digitalized image is divided into connected segments which are examined by one or several neural networks. The sample is classified as pathological if cell types are present which do not belong to the type of sample or if structural cell or tissue changes are present. In a second method of examination the digitalized image is again segmented and the segments are examined for the presence of a cell object. This is followed by an examination whether the cell object is an individual cell or a cell complex. In a third step of the analysis it is determined whether the found cell object is located on one of the image borders. If this is the case then a further image is recorded in which the found cell objects are completely included. Finally the segments in which cell objects have been detected are analysed at a higher magnification.