Gotcha. Are you looking to see changes in arterial tone and/or blood volume? We don't have any datasets and haven't worked with any PSG/HSAT systems that output the data beyond the Pleth signals you'll find in the datasets we've discussed already.
We are doing a number of WatchPAT (https://www.itamar-medical.com/watchpat-main/) projects, though posting those data are years away since the studies are still active. Here's a screenshot from a test recording, just curious whether this might be getting closer to what you're hoping to find. Unfortunately there wouldn't be any EEG alongside the WatchPAT studies.
The MESA EDF files should contain a "Pleth" signal. I saw this signal in the first 3 files I opened, e.g.
The review committee typically finishes reviewing data requests within 2-3 weeks of their submission.
I am not aware of any datasets like that available right now (here or elsewhere). We are starting some collaborations with researchers who do in-lab studies related to circadian rhythms. Some of these protocols involve forced desynchrony and varying light exposures. I don't have a firm idea about when any of these data might be posted, but you might want to check back occasionally. Thanks and good luck!
The scoring manuals should describe the scoring procedures for different datasets, e.g.
There isn't a "final diagnosis" per se since the sleep studies for most datasets were conducted for research purposes, so there wasn't an MD interpretation at the end. Users could try computing their own diagnosis variable from the indexes of interest and questionnaire data.
SHHS was an observational study, so participants were not explicitly chosen as case/control (or unhealthy/healthy; with a certain disease or without). As such, I think my recommendations from the discussion thread you linked are still accurate. If you want to stratify by Apnea-Hypopnea Index (AHI) severity, then AHI_A0H3 is an AHI variable that is very commonly used (in research and diagnostic settings). We create a lot of variations of the AHI for different research purposes, but AHI_A0H3 is arguably the "most important".
Thanks again for checking out the resource!
For most datasets you should be able to get an idea of the number of datasets/records by reading the documentation, dataset introduction pages, and looking at variable distributions. For instance, in SHHS, you could look at these links to get a sense of how many subjects participated at each visit.
Thanks for your interest in the site and good luck on your sleep stage classification project.
I just tested this from home and was getting around 50-60 Mbps, which is about what I would expect given my typical connection speeds from this location. I hope your IT team is able to able to help, otherwise it may just be some poor routing or something.
There should be no restrictions. We have not had reports from other users about slow download speeds.
How are you downloading files? Through the NSRR gem or through your web browser? I will make a note to test downloading from an external location to see if I am also affected.
The scorer reviewed mesaid = 2 and noted that there is artifact on the ECG channel that starts in epoch 156 and ends in epoch 158. The HRV reviewing/editing program (Somte) presumably censored the Rpoint data from those segments. I think it's reasonable to assume this (i.e. excluded ECG artifact) will be the common thread across other missing segments/areas of the files. Note that the ECG beats were only reviewed/analyzed in the sleep period between sleep onset and lights on.
Hope this helps in your work!