data2health / ehr2hpo.prj Goto Github PK
View Code? Open in Web Editor NEWConversion of EHR data (such as LOINC) to HPO codes
Conversion of EHR data (such as LOINC) to HPO codes
Implement the algorithm described here and test it using artificial or genomic data as preparation for the CTSA analysis
Watkinson J, Wang X, Zheng T, Anastassiou D. Identification of gene interactions associated with disease from gene expression data using synergy networks. BMC Syst Biol. 2008;2:10.
Anastassiou D. Computational analysis of the synergy among multiple interacting genes. Mol Syst Biol. 2007;3:83.
Add blog from npj
We will write up documentation that will live on the main HPO website that will explain LOINC and the LOINC2HPO project. This will be linked to the new pages on the HPO website. It will also use material from #9
Can we make a list of LOINC terms that were not mapped at one real-life CTSA, and look at the most common ones to identify any potential systematic reasons for failure? That is, if we are expecting the wrong codes or something like that, then the failure might not be because of incomplete curation (we know about that) but because of a logic problem in our code that we are not aware of. This analysis will try to figure that out by looking at the top 250 most common failures. If possible, it would be good to share this in a de-identified way with the entire team.
We should try to identify at least 2 CTSAs (and better 3-5) that would be willing to implement a LOINC2HPO analysis project
test issue :-0
Automatic ingest script.
One possibility is to add a "tab" on HPO term pages (next to the disease and gene tabs) and call it "Test"
Link out to LOINC like this: https://s.details.loinc.org/LOINC/2345-7.html?sections=Comprehensive
We will probably want to indicate if the test is directly related to the HPO term or if it is an inherited (true path rule) association.
Ask Julie for advice about look and feel.
Examine current status of these ontologies with respect to HPO mappings
Can we move this to a production environment, keeping all of the data within the firewall, as a next step?
right now the eval plan is largely the same as the deliverables. It would be good to have some evaluation measures.
What functionality will there be and how will you know the app has achieved it?
How will you measure user satisfaction or query answering capabilities?
What constitutes success?
Here is a simple example of an evaluation plan with more of these sorts of items, you could copy the template: https://github.com/data2health/Operational-architecture/blob/master/evalulation.md
Here on the demo linked from labs, https://octri.ohsu.edu/hpo_on_fhir/, the language needs to change to either:
The text in the header is easy to miss. Needs to be more Plug-and-Play whether hyperlink (eg. https://hpo.jax.org/app/ or button eg. https://alpha.monarchinitiative.org/)
Moreover, searching by patient is the least interesting form of what the LOINC2HPO makes possible. We should be able to search by phenotype, preferably one shared by a bunch of patients. Ultimately this should be possible using autocomplete, however, for now, some hardcoded examples (prepopulated per above) would be fine.
Not funded by CD2H but related.
The HPO team has been contacted by a European radiology project with lots of data on radiology findings with the goal of ammping this to HPO Terms
Is it allowable for us to add the LOINC mappings to the HPO website? Ask Daniel V exactly how we need to show the license
In the original project, we looked at a cohort that was enriched in but not exclusively composed of asthma patients. This type of cohort would be good to have for a larger project.
the RPPR doc https://github.com/data2health/ehr2HPO.prj/blob/master/RPPR.md
isn't consistent with the milestones and deliverables on the ReadMe.
Also please add dates to the timeline.
thx.
Determine new authors list
Finish revising text
Finish response letter
test
should this repo/project be renamed ehr2HPO rather than HPO2ehr?
And any reason it is private? the tracking repos must be public. Ideally the code is too ;-).
We will write up a report with an analysis of the performance of the HPO on FHIR app at OHSU and diseminate it on the CD2H website.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.