We are building the world's largest health graph
In our last post, we introduced the latest thing the Data Science team is building for PokitDok: the PokitDok HealthGraph. Today, we'd like to tell you why we are excited about this technology and give you a closer look into the connectivity of the PokitDok HealthGraph.
The top 3 things we love most about the PokitDok HealthGraph
1. Healthcare is naturally connected - let’s look at it that way.
Healthcare, at its core, is about people, procedures, and places, and the relationships between them. Modeling these sorts of relationships - or any sort of relationship between entities, really - is where graphs excel. By modeling the data more organically, we spend less time figuring out how to hammer the data into tables, and more time exploring and resolving questions that are important to you, our customers.
2. HealthGraph = Healthcare Transparency
Graph analytics are able to bring truth and transparency to ambiguous systems. For example, Facebook used their social data and graph theory to disprove the 6 degrees of separation myth. Right now, there is - and this won’t come as a shock to anyone - very little transparency in the healthcare industry. It is our hope that modeling health data in this way will lead to similar insights, in addition to added transparency, increased quality of service, and decreased costs for everyone.
3. What we already see looks awesome.
We have extracted a subgraph of the PokitDok HealthGraph and want to share it with you. The picture below shows all of the providers within the state of Washington and the properties (similar features) shared by the doctors (green nodes) and healthcare organizations (bright pink nodes).
The size of the node corresponds to its degree (the number of edges coming in and out of it). In layman's terms, bigger dots have more connections than the smaller ones. For example, some of the larger bright pink nodes correspond to large organizations of providers, like hospitals. The large blue nodes show densely populated cities with many practicing healthcare professionals.
For more in-depth analysis, let’s focus on one particular city. The image below focuses on Spokane, in purple, and shows the different types of individual doctors (green) and organizations (bright pink) within that city.
These visualizations are just the tip of the iceberg. With the PokitDok HealthGraph, we are able to dive into the data and begin to look at what is truly happening within the healthcare system. This directly translates to more accurate and transparent information which will in turn lead to more informed healthcare decisions for you, the consumer.
Hopefully this post gave you a brief introduction for why we think putting health data into a graph is such a good idea. In a future blog post, we'll be doing a deeper dive into the sorts of analytics the PokitDok HealthGraph enables. We'll further examine the connectivity of the PokitDok Health Graph, and see how the networks between physicians and other healthcare providers interact, as well as look at how the PokitDok Health Graph compares to other kinds of networks. Lastly, we will be showing how we are adding all of this information with the endpoints (claims, benefits, price transparency) from our platform APIs into the world’s largest HealthGraph.
- Snake_Byte #28: Term x Document Matrices via Sklearn's DictVectorizer - April 21, 2017
- DC Blockchain Codeathon - April 11, 2017
- Snake_Bytes #5: Vector Distances - October 12, 2016
- TinkerPop and Titan from a Python State of Mind: A Tutorial via Docker - December 15, 2015
- Online Recommendation Systems: Getting Personal with Gremthon - Part 2 - April 14, 2015
- Online Recommendation Systems: What it Really Means to be 'Somebody Like You' - Part 1 - April 7, 2015
- The top 3 things we love most about the PokitDok HealthGraph - January 27, 2015
- The PokitDok HealthGraph - January 20, 2015
- Snake_Byte #14: Writing Made Easy - January 6, 2017
- Top 100K PokitDok Providers - November 30, 2015
- Visualizing Medicare Physician Co-Occurrence and Payment Data Geographically - Part 2 - July 14, 2015
- Visualizing Medicare Physician Co-Occurrence & Payment Data Geographically - Part 1 - July 9, 2015
- National Provider Data Revealed - A Choropleth Map Analysis - October 9, 2014