Saturday, July 1, 2017

SKED -- What we might do with all this health data!

Data data data data...

...what are we going to do with all of it? AND how are we going to get all of it to the ultimate goal -- early automated disease prediction?

Scientific Knowledge Extraction (from) Data (SKED) is one pipeline that you can read about in BioRxiV here.

It is a little vague, but I think it has to be. We're going to have to approach the healthcare big data problems with big ideas and plans set out way in advance. The main point of the paper is the reduction of input data down to "data primitives". They don't focus on this a ton, but the primitive data is going to also have to drop patient identification data that will need re-associated if patient diagnoses come into play. For scientific inquiry into the sets, we won't have to worry about that part.

Maybe these ideas are obvious, but I thought it was an interesting short read.

1 comment: