Strategies to improve quantitative assessment of immunohistochemical and immunofluorescent labelling

Binary image thresholding is the most commonly used technique to quantitatively examine changes in immunolabelled material. In this article we demonstrate that if implicit assumptions predicating this technique are not met then the resulting analysis and data interpretation can be incorrect. We then propose a transparent approach to image quantification that is straightforward to execute using currently available software and therefore can be readily and cost-effectively implemented.