Overview

ChRIS

At the FNNDSC, our team (the Advanced Computing Group) has developed a container scheduling platform called ChRIS. What is ChRIS? Well you could ask chatGPT "Tell me about the ChRIS Research Integration System at the FNNDSC". When I tried this, chatGPT said

"The ChRIS Research Integration System (RIS) at the FNNDSC is a component of the ChRIS project that provides researchers with a unified, web-based interface for accessing and integrating data from multiple sources, including medical imaging data, patient records, and other clinical data. The ChRIS RIS supports data integration through a combination of APIs, data connectors, and data pipelines, and provides tools for data visualization, analysis, and collaboration. The ChRIS RIS is designed to help researchers effectively manage and analyze large amounts of complex data, and to facilitate collaboration among researchers, clinicians, and data scientists working in pediatric research. The goal of the ChRIS RIS is to advance the field of pediatric research by enabling researchers to better understand and address the complex health challenges facing children today."

Close, but it got a couple of details wrong. You can try the actual ChRIS site here -- it's probably better: this is better than chatGPT.

Ok, so what's this visualization thing you want us to look at?

Well, first things first. The ChRIS system keeps track of medical image analyses in things we call "Feeds". The first link below shows a Feed Table, or a record of all analyses performed in the system. When you click on a Feed, you get a detailed view of that analysis -- see the second link. Notice the tree representation in the top left? That shows the logical relationship between all the compute nodes in the experiment.

As you can see in the second link, the layout of all the compute in the tree conforms to a directed acyclic graph, or DAG.

Notice the 3D button in the tree view? Well, if you hit that you see what the third link shows. Pretty cool, huh?

  1. * > ChRIS Feed Table <
  2. * > A 2D Compute DAG <
  3. * > Now in 3D! <

Think outside the box!

Hopefully this might inspire some ideas about how to think about expanding on the current visualization. What skills might be useful to have?