dwongh / openscore Goto Github PK
View Code? Open in Web Editor NEWSoftware for viewing and reporting EEGs according to the SCORE standard.
Software for viewing and reporting EEGs according to the SCORE standard.
This would give a clearer indication to the user that the context has shifted to the new recording.
Use directories with EEG and corresponding text reports to autopopulate the fields in the score report. The text from the report is currently loaded into the patient history box, but we need to remove prevbious diagnosis (for teetacsi), and potentially filll other fields if we can identify them
File -> Load EEG allows me to select a folder, but then I get the error message below.
'MainWindow' object has no attribute 'load_eeg'
EEG recordings may have patient any of sleeping, drowsy, awake throughout the recording and so should be able to select more than one
If a recording is 'Normal' the interpreter may go to the 'diagnostic significance' tab and see than 'Normal' is already selected (by default) then move on. This causes the diagnosis to be recorded as 'null' rather than 'Normal'. Is it possible to make these radio buttons behave like tristate checkboxes?
Several options given in drop down menus are likely missing correct or standard medical terminology. Need to review and identify these placeholders and update them if necessary.
The link in the readme yields '404 not found'. Is it a private repo?
Has this version of EDFBrowser referred to above been modified in some way that means that the official version doesn't work with OpenSCORE? If so, the README should make that clear.
Units for all 'frequency' and 'amplitude' fields in Findings should be indicated (e.g. in the static text) as Hz or microvolts.
In the first 'amplitude' drop-down box, units are currently written as millivolts ('mV'). These should be microvolts, ideally written as V. If it's not easy to get the Greek mu to appear, 'uV' is fine.
EDFBrowser stays open in separate process when OpenSCORE is closed. Need to add a close method to OpenSCORE to do clean up.
Score files currently store 'false' or 0 for any checkboxes that were not ticked. This could cause confusion between whether something has been confirmed as false or simply not considered. Would it be possible to use tristate checkboxes instead?
Under Recording Conditions, date & time of recording and latest meal are auto-populated with the current time. This creates a risk of capturing false data if the user fails to overwrite these values. Ideally, they should be left blank. If that's too much hassle for now, they could be auto-populated with a date in the future (e.g. by one year) and then a simple check against this date can be used to prevent future dates from being saved.
I couldn't find which package might need to be installed for this to work. Is it a custom module that hasn't been pushed to the remote?
I tried skipping this import and swapping 'QMainWindow' in for 'MainWindow' in the code. That ran, but just opened a blank window.
Creating the mirror directories currently relies on the structure of the TUEG dataset. Specifying paths to edf recordings which are not in the same structure as the the TUEG dataset will fail
Hopefully OpenSCORE will become useful to a fairly broad community of researchers or even clinicians. At the moment, some features such as eye-tracking are not needed by general users. The complexity of OpenSCORE should be minimised to ensure its relevance. How it would it be to separate 'general purpose OpenSCORE' from 'TEETACSI OpenSCORE', e.g. by having a separate repo for TEETACSI, using this OpenSCORE repo as a submodule?
Features to consider separating, in descending order of priority:
Perhaps it's best not to act on this just yet, to postpone introducing the dev complexity of managing and documenting two repos. Let me know how much of a headache you think it would be though.
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.