If I have understood right, hypopneas were scored from airflow, thorax or abdomen signals. Nowadays, AASM rules state that hypopnea scoring should be done from airflow signal, which was not the case while SHHS datasets were scored.
RESPSCCH variable has some information from SHHS1 measurements, which signal(s) were used at hypopnea scoring, but almost half are still unknown, and for SHHS2 dataset there is no such variable.
I would like to know if there is some variable, or other way, to know which signals were used at hypopnea scorings.
I am matching my own desaturation scorings to respiration events (in profusion.xml files), and calculating AHIs (and ODIs) and our research team's own parameters, which include duration and/or area calculation of respiratory events. But if I use scorings from thorax/abdomen belt and calculate related area/duration in airflow signal, this could cause problems/inaccuracies in some cases. Especially if no clear event is seen in airflow signal. Or at least it would be difficult to state that desaturation-respiratory event matching was performed according to AASM rules.
I hope you guys have some idea how to get around this problem, or some suggestion how to identify if event scored in thorax/abdomen signal could have been also identified from airflow signal.
Yes, you are correct that hypopneas were scored from the airflow, thoracic, or abdomen signals for SHHS. I don't believe there is a way to identify which signals were used for hypopnea scoring for SHHS2. Taken from the manual "Variation in signal amplitude on airflow, thoracic and abdominal channels: In some studies, information appears qualitatively different from different channels. Scoring will be done from the channel that correlates the best with the changes in the oxygen saturation. If there are no changes in O2 saturation, scoring will be done from the channel that shows the clearest amplitude variation. In cases where the thermister varies from the inductance channels, and the inductance channels appear mostly artifact free, the inductance channels will be used for event identification and classification. When in doubt, use data from the inductance and saturation channels to identify/classify events." The RESPSCCH variable was not collected for SHHS2. I am not aware of any other variables which would provide the information you seek.