Frequently asked questions (FAQ)

What is computational pathology?

Computational pathology is a new phrase that we thought of to describe the interface between technologies used in digital pathology, and algorithmic strategies used to solve computational problems in other areas like bioinformatics, computational biology, machine learning, or more general imaging communities.

Two great examples of challenges that we would classify under "computational pathology" are the mitotis cell count competition (ICPR 2012/2014) and the recently launched challenge on cancer metastatis detection in lymph nodes (CAMELYON16)

Do I have to do research in a medical background?

No, you don't. It's just that that's where the obvious applications are. We've seen great examples of innovative algorithms over the years: the automated annotation of C. elegans micrographs is an example. The 3D-reconstruction of D. melanogaster neurons based on confocal microscopy another. We welcome applications outside medicine, because we feel the technology of whole slide imaging definitely has potential to be used outside clinical pathology.

Say that I'm an NGS scientist; how can digital pathology help me?

In pathology there's a saying: "tissue is the issue". This means that we feel that morphology still matters in an NGS world. The idea is that if you were to individual cell sequencing, you would get a better understanding of the results by also taking into consideration the relative orientation of the cells from which you extracted the genetic material or transcripts from. Digital pathology techniques can also assist you to help you identify the most interesting cells to apply *omics techniques on in the first place. When you combine this type of methods with NGS, you're a computational pathologist.

Can you provide some examples in the recent literature of computational pathology?

Certainly. A list of relevant literature is provided in a separate document

What kind of freely available software is there to start working with this type of data?

Computational pathology is not about tools; it's about the applications. That being said, we're certainly willing to point you in the right direction to get started:

http://www.openslide.org

http://www.openmicroscopy.org

http://imagej.nih.gov/ij/

http://free.pathomation.com

http://www.cytomine.be

http://cellprofiler.org