NSRR staff
Boston, MA
0000-0002-0506-8368
Top Topics
Recent Topics
Hey,
Please try the SOF Online website to download other SOF data that can be linked with the polysomnography data here on NSRR.
Hi,
Can you please provide output for these commands?
ruby -v
nsrr -v
Yes, you can apply for multiple datasets at once. We understand that deep learning applications will require a large amount of data.
I don't fully understand your question. Could you please restate?
It would not occur to me to group our datasets and make conclusions about our datasets based on "these datasets are non-commercial use only" or "these datasets allow commercial and non-commercial use". The decision to share for commercial use or not is a separate decision made by each dataset owner.
The number represents the dataset version. You can see the full CHANGELOG here - https://sleepdata.org/datasets/shhs/files/m/browser/datasets/CHANGELOG.md?preview=1
The files in the main /datasets folder (https://sleepdata.org/datasets/shhs/files/datasets) represent the latest/newest version of the dataset. There is more information about the dataset here - https://sleepdata.org/datasets/shhs/pages/04-dataset-introduction.md
Mike
Thanks for raising the issue. The CHAT SaO2 values > 100 is discussed on these pages:
Note that only one (1) participant has values exceeding 100 during sleep time. Wake time was not always reviewed for signal artifact.
These data are provided "as-is". We leave it up to the user to decide how to handle (e.g. exclude portions of signal(s), recode values, etc.)
Good luck and thanks for checking out the site!
Requests are typically reviewed and responded to within 1-2 weeks of submission. You should hear back soon. Thanks for checking out the site!
Hey - sorry, we've been having a hard time getting in touch with them as well. We will eventually remove the dataset if we're not able to reach them for a discussion about sharing their dataset.
The Learn dataset contains some public EDF/XML pairs: https://sleepdata.org/datasets/learn
Unfortunately we no longer have anyone on the team that supports the SpectralTrainFig tool. The Luna tool does a lot with EDFs and EEG data and is actively being developed.
Hey Ankit - thanks for using the site and for bringing this to our attention. Your finding is correct about slptime. The documentation in the original SHHS datasets about this variable was not clear, so I have identified it for removal from future versions of the dataset (https://github.com/nsrr/shhs-data-dictionary/issues/1107). slpprdp is typically our preferred variable for total sleep time.