Healthcare at rstudio::global(2021)

If you'd like to start a new discussion under Healthcare, go to #industry:healthcare , and click "New Topic".
To follow new posts, be sure to at least click "Watching First Post" on #industry:healthcare, and click "New Topic".
image


Recap

Here's a recap of some of the activity related to Healthcare at rstudio::global(2021)


What do you suggest following and reading?

Shannon Hagerty to Healthcare:
Hi All, I'm curious what are your favorite resources for learning more/ seeing examples of R in healthcare. I'd love to see any github repos, books, websites, blogs, twitter accounts you'd recommend!

Zoë - The NHS-R Community GitHub has a lot of resources including the training slides we use and presentations at the conferences: https://github.com/nhs-r-community Twitter is: @nhsrcommunity

Adam - Here is an example that characterized COVID-19 patients in the OHDSI network: https://data.ohdsi.org/Covid19CharacterizationCharybdis/. Network studies like this are distributed as R packages and can run on any dataset in the OMOP common data model format.

Peter - WIP, but this one is coming along...https://bookdown.org/pdr_higgins/rmrwr/
Trying to get interactive features nailed down first, then roll out more content

Aedin - The Bioconductor community have recently started creating books, The osca book is a great resource for single cell data http://bioconductor.org/books/release/OSCA/

Daniel this is my dissertation project: https://ds4biomed.tech/
for people making new materials, may I point to the "who is this book for" section for the learner personas :slight_smile:


Poll - As an R user in healthcare, I am most interested in...

Jason Milnes to Healthcare - January 21st, 6:14 AM

Let's kick this off with a question and feel free to share more in the comments!

As an R user in healthcare, I am most interested in...

Q&A - What is the top community open-source effort that could help further data science in healthcare?

Phil Bowsher - Is open data a challenge in this space? I know AI usually needs large datasets that aren't always available, especially across organizations or combined. I've seen efforts like iedb.org and other data-sharing initiatives such as ADNI and the Cancer Imaging Archive. Is this a top challenge for this group?

Daniel Chen -January 21st, 12:21 PM
ohdsi: https://www.ohdsi.org/

Cameron - OHDSI looks like a great community effort. Wondering if omics data and other non-structured data is standardised/planned to be?

Peter - Cameron, I think a lot of that is done via Bioconductor

Marten - FHIR

Aedin Culhane - Cameron I work with Bioconductor and recently have also started working on a few COVID19 related OHDSI projects, I am interested is seeing how we could bridge the clinical genomics space. There is also the CDSIC eTRIKS community but that isn't in R (as far as I know). The OHDSI recently published a paper expanding their oncology ontology, and I supposed this could be expanded to include 'omics ontologies but I think there is a lot of work to be done in this space.

Adam - Cameron MacPherson Here is a discussion about omics data on the OHDSI forums. It has not been integrated into the common data model and most EHR datasets do not have it to my knowledge. Seems like an area that will become more important as omics data gets into the EHR.

Cameron - Thanks for all the answers! Sorry to move off topic. There's a lot of movement in this space, so I guess one strategy might simply be to wait as tech has transferred into the clinic as @Adam Black suggests, like with some deep immunophenotyping and genomics platforms. The problem for me here is "data communication" as this step usually signals the end of transparency essential to analytics (which is why these open efforts are :ok_hand:). I'll introduce myself outside this thread in case we continue this discussion :slight_smile:

1 Like