Fig. 2From: Clinical application of the HM-1000 image processing for HER2 fluorescence in situ hybridization signal quantification in breast cancerLinear regression model used to estimate the optimum background threshold. The preset linear regression model was used to determine the optimum threshold, then the reverse threshold was determined and the >95% correlation between the optimum threshold value and the mean (standard deviation) of the image was calculated. The optimum threshold value of each image was estimated using this regression modelBack to article page