Color standardization in whole slide imaging: a method to reduce color variability

Color standardization in digital pathology is essential for ensuring consistency across Whole Slide Imaging (WSI) scanners. This study utilizes color data from multiple WSI scanners captured from a slide containing 55 color patches with histology stain spectral characteristics alongside a proprietary algorithm to generate International Color Consortium (ICC) profiles. By transforming scanner-derived color values to align with their true spectral counterparts, the method enables effective color standardization. The color gamut of various scanner models was compared to the true spectral values both before and after standardization, as well as to the standard Red, Green and Blue (sRGB) gamut commonly used in display technologies. Results show that WSI scanners apply significantly different linear and tone curve profiles, as well as introducing variability in white and black point calibration. The proposed ICC-based standardization method substantially reduced discrepancies between scanned color values and their true spectral equivalents. This approach demonstrates the potential to mitigate scanner-induced color variability within digital pathology workflows. As artificial intelligence becomes increasingly integrated into pathology, addressing color presentation discrepancies between human observers and machine learning algorithms is critical. Application of standardized calibration would ensure consistency of data used for accurate diagnostic methods leading to reliable use for scaling of digital pathology. The uniquely large dataset presented here reveals how different image sources can generate significantly different outputs. Industry best practice would be to have all scanners calibrated by the same standard method and quantifiable accuracy, leading to real standardization rather than the variety of approaches this paper has revealed.