dcan-labs / bids-tryworks Goto Github PK
View Code? Open in Web Editor NEWA graphical tool that indexes a folder of dicoms and provides the user with an interface to convert those dicoms into BIDS.
License: BSD 3-Clause "New" or "Revised" License
A graphical tool that indexes a folder of dicoms and provides the user with an interface to convert those dicoms into BIDS.
License: BSD 3-Clause "New" or "Revised" License
Session Detail table at bottom of page should update when you click on subject
This feature is currently implemented, but it's bugged majorly in that it displays every single session for every single subject. Bug Greg Conan to fix.
See above question.
It is unclear what such does when one enters items into both Study Description & Study ID. It seems that search will return anything that matches either the Study Description OR Study ID fields. But this needs to be tested, confirmed and potential for feature addition.
grab pIDs on dcm2bids conversions. update page with basic progress ("Working on conversion 12 out of 20"). show "user" and "time started".
Line 359 in aa52ec3
In order to auto-populate the dcm2bids conversion builder we'll store user selections in a dictionary and look them up each time the conversion builder form is loaded.
AKA search the dicom information, match it to BIDS specs, populate dropdown with the closest match - provide "other" manual category label as well
Currently the docs folder contains some information that is only relevant (but possibly errant since this code was refactored from bidsgui2 into bids-tryworks) to the setup and installation at OHSU. These files need to be updated/removed. see /docs .
One thing that we never do, but that throws errors when we process, is write the dataset_description.json that is supposed to live at the top of the bids-input. It only has 2 required entries: 'Name' (the name of the dataset) and 'BIDSVersion' (the version of BIDS to which it conforms - currently v1.2.0). It would be awesome if the BIDSGUI would write that file when it runs dicom2bids. (I think dicom2bids should do it, since it knows what version of BIDS it is using, but it doesn't do it.) The spec is here:
Ideally there is only one bidsgui2 install per server. Currently there are 2 installed on rushmore, but I believe we can get this number down to one.
Steps:
This is both a question & a feature request. It is important to determine if the dicom to nifti conversion will continue to take place if the browser is shut down.
The advantage of this software is especially apparent when data for larger groups of subjects. This would be hindered by requiring that the browser remain open in order for the conversion to complete. This issue is of importance due to the increased remote nature of our working environment.
tryworks removes the directory containing tmp_dcm2bids by default. This should either be removed entirely, or made to be an option.
AttributeError resulted after final stage of process ("convert"). No screen shot was captured. The error did occur in line 635 of /dicoms/views.py.
It would greatly improve usability if the search path (or dicom directory) were to be user defined in the GUI itself. As it stands now, there is zero documentation on how to use this software for other studies/projects. The user would have to edit a hidden file that is barely mentioned in the documentation. This will be further complicated when attempting to convert dicoms for multiple studies.
Optimal use would allow the user to adjust the dicom source directory inside the GUI environment.
The "CONVERTED_FOLDER" in the .env file cannot match the DESTINATION_DIR as set in the GUI itself. When they match, the program runs, but then deletes the converted files in the CONVERTED_FOLDER. Given that it matches the DESTINATION_DIR, the outputs are then deleted.
Prevent user from clicking anything on the conversion page after submission.
2 Questions:
Is bids-tryworks capable of running on exacloud?
Is bids-tryworks capable of writing converted files on exacloud (from rushmore)?
Make a single launch command which:
manage.py runserver <port number>
firefox localhost:<port number> &
This will avoid confusion about which port to use, and will also save the user the hassle of running the server in 1 command window and a browser from another.
In order to process subjects with the abcd-hcp pipeline, I need to have the subject ID and session dates.
Currently the BIDS GUI does not report session in a way that allows the user to record the appropriate subject ID with their session date (either in a .csv file or any other method). This is the format I use to process the data:
${sub-id},${ses-date}
The gui is able to display the date of the scan after I search for the subject ID's, and before I make the conversion json and niftis.
Having this resource will allow me to start processing these subjects in a more efficient manner without having to hunt down the dates.
Another change suggestion is to be able to modify the session dates along with the sub ID's
Conversion is halted/errors out when executing dcm2niix command on 099_Phoenix_Document. This error occurs because the Phoenix document is a Siemes proprietary file type to allow transfer of protocols from one scanner to the other. It does not contain DICOM files and cannot be converted. This is a bug that halts the conversion of any data that contains such a directory.
Seems like the error occurs in dicoms/views.py due to Session.object thats created in dicoms/models.py. Further details are in a screenshot that contains non-public data.
Presently all image/modality types are listed in the same drop down menu with no delineation between anat, func, dwi, fmap, or beh. This can lead to confusion the part of the user, so instead of grouping all of these in one drop down they should instead be separated into their categories some how. Whether that's a nested drop down or some other method.
This is mostly just to make github not be annoyed at our bad security practices. Also, note this application is not meant to ever be hosted online. It's meant for use on a local machine.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 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. 📊📈🎉
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 ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.