OHDSI Home | Forums | Wiki | Github

Feedback wanted! Tool to visualise multiple codes prevance in a OMOP-CDM

#1

Hello all,

We have a PhD student in data visualization who is developing a tool to visualize data in the OMOP-CDM.

She needs to conduct a few user interviews on this tool, and we would like to invite you to participate.

In short, the tool will take as input a list of concept IDs and plot together the over-time prevalence of the concepts.

Similar to what Atlas can do for individual concepts, but this tool will show several concepts, including non-standard concepts, on the same plot.

We see several use cases for this:

  1. Plot all the non-standard concepts mapped to a standard concept to understand how vocabularies have evolved in the database. For example, Allergic asthma SNOMED code 195967001 being first mapped as the ICD9 code 493.9 and then ICD10 code J45.9.

  2. Plot all the drugs under an ATC code to understand how drug products have evolved over time.

  3. Plot a list of concepts suggested by PHOEBE to understand how the prevalence of these concepts has evolved over time.

Please let me know if you are interested in participating in this user interview or if you have any feedback on this tool.

1 Like
#2

Hi @Javier, sounds very interesting and I would be happy to participate in an interview.

If I am understanding you correctly, it will be like the Achilles concept drill-down, but then displaying multiple concepts at once. Note also that the stand-alone Achilles viewer has been re-implemented in Ares.

#3

Hi, @Javier

Very interesting, I would like to participate in an interview.

#4

Thanks @Javier , I’d love to see your team’s progress and happy to be part of interview process.

I’m interested to see how it compares with Chapter 15 Data Quality | The Book of OHDSI (Figure 15.2), which provides monthly prevalence over time sparkline for a set of concepts/source codes as a quality tool. I have been surprised and disappointed that this visualization hasn’t become a more embedded part of our processes, as it can be quite helpful to identify potential concerns when building conceptsets.

#5

Hi @MaximMoinat,
I’m the student Javier mentioned - happy to hear you’d be interested in doing an interview! I’d love to set up a time to meet. Just let me know your email and we can coordinate something.

#6

Hi @zhuk,
I’m Veronica, the student Javier mentioned. Happy to hear you’d like to participate in an interview. If you’re still interested, let me know your email and we can set up a time to discuss the tool further!

#7

Hi @Patrick_Ryan,
I’m Veronica, the student Javier mentioned. Thanks for sending along this reference, I haven’t looked into it before and it seems like it could be relevant. If you’re still interested in doing an interview, just let me know your email and we can set up a time. Thanks!

#8

Hi @Javier and @Veronica_Lorenzini I realise this thread is a few months old now but I was wondering how your visualisation project is progressing? I’m looking at creating some visualisations of OMOP data and wondered if what I am attempting to do is in any way similar to your research. I am trying to take a cohort of patients, eg epilepsy patients, and visualise the type and volume of data contained in our OMOP database that relates to their care. It sounds like your visualisations are related to concepts and not patients but there might be some overlap. I’d be very interested in discussing this with you if you’re open to that?
Fran

#9

hey, @FranB thanks for your interest,
Attached
is the thesis of Veronica

Initially, this is to look at the concepts counts and prevalences in the whole database.
However, we are thinking that the same can be applied to a cohort.

#10

umm sorry, It was to big to upload,
here aTemporal link to it
https://filesender.funet.fi/?s=download&token=20f97873-6e12-40cb-9a9d-a98da0f609d7
i will add the official link once it is public

1 Like
#11

Thanks!

t