NSRR staff
Boston, MA
0000-0002-0506-8368
Top Topics
Recent Topics
Thanks Sharon - it's a good question and has come up for discussion in the NSRR team before. At present we don't have any plans to extract PSG metrics from the TSV annotations.
The NCH publication (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9296671/pdf/41597_2022_Article_1545.pdf) has this to say:
Sleep studies were annotated in real time by technicians at the time of the study, and then were staged and scored by a second technician afer the study was completed. Technicians annotated studies using a combination of free-form text entries and selections within Natus Sleepworks. Technicians tried to identify all events of interest, however each technician may have their own style of text annotation. Due to the variability in sleep stages in children, NCH does not use automatic scoring of sleep stages. All sleep stages were manually scored by a technician and then verifed or changed by a physician board certifed in sleep medicine. Sleep studies were manually downloaded and converted to EDF+format between May 2019 and Feb. 2020 using Natus Sleepworks version 9. Any gaps in the time-series data were padded with zeros as part of the conversion. Te specifc acquisition equipment, setup, and montage all followed standard care protocol at NCH. While changes may have been made to some studies, the NCH protocol for PSG is in accordance with the rules and technical specifcations recommended by the American Academy of Sleep Medicine10,11. Standard channel names are used and documented in the header of the EDF fles, allowing inference of the montage.
Sleep studies were annotated in real time by technicians at the time of the study, and then were staged and scored by a second technician afer the study was completed. Technicians annotated studies using a combination of free-form text entries and selections within Natus Sleepworks. Technicians tried to identify all events of interest, however each technician may have their own style of text annotation. Due to the variability in sleep stages in children, NCH does not use automatic scoring of sleep stages. All sleep stages were manually scored by a technician and then verifed or changed by a physician board certifed in sleep medicine.
Sleep studies were manually downloaded and converted to EDF+format between May 2019 and Feb. 2020 using Natus Sleepworks version 9. Any gaps in the time-series data were padded with zeros as part of the conversion. Te specifc acquisition equipment, setup, and montage all followed standard care protocol at NCH. While changes may have been made to some studies, the NCH protocol for PSG is in accordance with the rules and technical specifcations recommended by the American Academy of Sleep Medicine10,11. Standard channel names are used and documented in the header of the EDF fles, allowing inference of the montage.
I don't see any other notes/documentation on this end about the scoring rules/guidelines.
The primary NCH publication contains some information on sampling rates: https://pubmed.ncbi.nlm.nih.gov/35853958/
The total length of recording in the NCH Sleep DataBank amounts to 40,884 hours, where the minimum length of study is 3minutes, the maximum is 16.5hours, and the mean is 10.3hours. 94.85% of the fles contain between 8 and 12 hours of recordings, and the patients slept for a subset of those times. Users of the dataset should take into account that the majority of the recordings (3,204) are collected with a sampling frequency of 256Hz, but 581 studies were sampled in 400Hz, and the rest (199) in 512Hz.
Thanks for checking out the site.
The dataset READMEs (e.g., https://sleepdata.org/datasets/shhs) and data request language (e.g., https://sleepdata.org/data/requests/shhs/start) stand-in as a license file, describing appropriate and inappropriate uses of the dataset. Datasets that have commercial use and/or subject type restrictions will state that in the README.
Thanks for visiting the site!
Thanks for the question. There hasn't been a "re-annotation" in the way you're thinking (i.e., re-opening and re-scoring of the PSG data); this is more of a mixing and matching of the originally scored respiratory events to better conform to more modern guidelines.
Thanks for raising this issue - we have realized this is a limitation in many of our older datasets that were scored in Compumedics Profusion. That software doesn't export lights off/on information into the EDF or default XML annotations.
For some datasets, such as MESA, you can find the lights timing variables in the CSV covariate dataset (in the /datasets folder). For instance:
In SHHS, I only found lights off (https://sleepdata.org/datasets/shhs/variables/stloutp).
In the future we intend to encode lights information inside of harmonized annotation files. Until then, please use the variables in the dataset!
Those variables have the same meaning in the non-randomized dataset as they do for the randomized subjects, so you can refer to the meanings in the variables browser. I don't see a ref6 variable anywhere, however.
https://sleepdata.org/datasets/chat/variables/ref5 (ethnicity)
https://sleepdata.org/datasets/chat/variables/ref8 (BMI)
https://sleepdata.org/datasets/chat/variables/ref9 (gender)
Hope that helps and thanks for using the site!
Hey - thanks for checking out the site! Unfortunately, you are right, we do not have LDL in SHHS. The lipid panel variables were "A variables", which means they weren't collected as part of the SHHS protocol, but rather requested from the parent cohorts separately (to be taken from separate research visits). Evidently, LDL was not in the "A variable" request list.
I pinged the STAGES team for input. I see the description in the MOP you're referencing. Note that the STAGES project ended prematurely so they may not have been able to complete all their original aims.
Thanks for using the site!
I heard back, unfortunately the position signal encoding information was not retained by the STAGES data contributor, so we can't be certain of the different position signal value meanings across the STAGES sites.
Thanks - if you have an algorithm to detect the EKG-related data yourself then I recommend using that. The information on the HRV page (https://sleepdata.org/datasets/mesa/pages/hrv-analysis.md) is all the information we have. I believe only the R-points in sleep were intended to be reviewed.