Should Digital Pathologists be selected on their Biological Visual Performance ?

As a Pathologist adopting new technologies such as Artificial Intelligence & Deep Learning it remains essential to be consiously aware of our limitations, especially related to visual and related cognitive functions. Often digital pathologists are asked to support computer scientists to develop new algorithms. These are often build by different pathologists in order to reduce the bias as much as possible. But what is no one is aware of the bias ? What if the algorithm has integrated the visual limitations of each of us ? Should we select our pathologists in function of their biological performance ? Should we develop algorithms to correct human limitations as well ? Personally, I do not have the answers to these questions, but I am convinced that awareness is the key for improvement and achieving excellence.

One of the most interesting biological imitations we have, is distinguishing color hues. It is not a trap but a real human biological limitation ! Not all individuals neither pathologists have the same ability to tell staining intensities or shades apart from each other. Therefore, performance can be measured through the Farnsworth-Munsell 100 Hue cooler Vision Test. The lower your score, the better color vision performance you have. Surely, you will need to calibrate your monitor before doing the test ! So, let’s do the test: there is no need to publish your results.