GiuliaDAP
commented
5 years ago Thanks for this comment. rri, t and tNN are time indices, indicating the time [seconds] related to a RR or NN interval. I’ll make sure the naming is consistent in all the functions and update description accordingly.
https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Tools/PRSA_Tools/prsa.m#L92
nn_win = rr( rri >= WinStarIdxs(i_win) & rri < WinStarIdxs(i_win) + windowlength );
rri is used to generate indices not as an index
On May 16, 2019, at 7:10 AM, embar- notifications@github.com wrote:
Second input variable for prsa.m function is "rri" is ambiguous and is used in two different contexts: index and time.
Cases where "rri" seems to be used as index: https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Tools/PRSA_Tools/prsa.m#L8 https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Tools/PRSA_Tools/prsa.m#L92
But "rri" is also re-used in context of time: https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Tools/PRSA_Tools/prsa.m#L60
Also "rri" comes from Main_HRV_Analysis.m (as "tNN"): https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L170 https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L140 https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L292 "tNN" created in RRIntervalPreprocess: https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L297 https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Tools/Preprocessing/RRIntervalPreprocess.m#L1 according to "cleantNN" description, it "tNN" (and then "rri") should be index: https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L297 Though computationally it could be time variable in seconds also. Anyway, source is "t" variable at "Main_HRV_Analysis.m", that also has ambiguous definition: https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L13
If these variables are time variables (units are seconds), then there is contradiction with cases where it is used as index.
If it is index please remove unit name "seconds" from Main_HRV_Analysis.m and include check whether it is integer. Also seems, we usually don't need it at all as input, because we can generate it internally, e.g. for Main_HRV_Analysis.m
if strcmpi(InputFormat,'RRIntervals') t=1:length(InputSig) end
in other cases it described not be used anyway: https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L14
Or to generate internally for prsa.m: rri=1:length(rr)
Please clarify these variables.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.
embar-
Second input variable for prsa.m function is "rri" is ambiguous and is used in two different contexts: index and time.
Cases where "rri" seems to be used as index: https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Tools/PRSA_Tools/prsa.m#L8 https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Tools/PRSA_Tools/prsa.m#L92
But "rri" is also re-used in context of time: https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Tools/PRSA_Tools/prsa.m#L60
Also "rri" comes from Main_HRV_Analysis.m (as "tNN"): https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L170 https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L140 https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L292 "tNN" created in RRIntervalPreprocess: https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L297 https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Tools/Preprocessing/RRIntervalPreprocess.m#L1 according to "cleantNN" description, it "tNN" (and then "rri") should be index: https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L297 Though computationally it could be time variable in seconds also. Anyway, source is "t" variable at "Main_HRV_Analysis.m", that also has ambiguous definition: https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L13
If these variables are time variables (units are seconds), then there is contradiction with cases where it is used as index.
If it is index please remove unit name "seconds" from Main_HRV_Analysis.m and include check whether it is integer. Also seems, we usually don't need it at all as input, because we can generate it internally, e.g. for Main_HRV_Analysis.m
in other cases it described not be used anyway: https://github.com/cliffordlab/PhysioNet-Cardiovascular-Signal-Toolbox/blob/aebdcc0d294158d6de1bff2088ca6e276d6ca97d/Main_HRV_Analysis.m#L14
Or to generate internally for prsa.m:
rri=1:length(rr)Please clarify these variables.