Aug 13, 2018 · 2 min read

An extension of the topic “How Blockchain can help with healthcare’s patient matching problem” by Bill Siwicki of Healthcare IT News

The topic I am expounding on is in reference to the section:

— Blockchain and ID challenges —

In the case of this Healthcare IT News article they fell into a ditch we are all familiar with.

They fell into the “well they won’t let me have the data and there is no other way to get it” trap.

The real question here is how do we identify patients with what we have and how do we make it relatively universal. Now you’ll see that I used the phrase “relatively universal ” because the data coming in needs to be well manicure.

We all know that identifying information or phrases can be encrypted with a known key and hashed can be used to assign a unique identifier. Websites that use a series of words in a particular order use a similar means of locking and unlocking.

The Really amazing things start to happen once you incorporate zero knowledge proof (ZKP) into the mix. Once a ZKP is introduced that unique identifier can be used as an ID without every sharing it at large, thus keeping it private.

The realization my team came to is that the tech is there for UPIs (universal patient identifiers) right now.

So how do we start to focus on ingesting data and predictively identifying bad entries and how would one accomplish something like that?

Our pocs indicate that Deep Learning / Machine Learning will be the key to mass identify bad entries and flag them for human intervention.

It’s closer to reality than most people would even imagine.


Written by


MedBlox is a decentralized, secure, and compliant means of exchanging electronic health information.