Open DominiqueMakowski opened 3 years ago
Hey @DominiqueMakowski,
Thanks for taking a look! I think you're right, that seems like the best way to integrate our peak-finding methods. I think an optional NK dependency in peakdet sounds great.
Unfortunately I'm swamped finishing up some last-minute analyses for my dissertation 😅 so I won't be able to get to these changes for a couple weeks.
Hello @62442katieb ! Trying to revive this issue. I'm going to also take a look at that for the rest of the codesprint. Let me know if you want to coordinate 😄
Hey folks, I took a quick look at the code here, and f I'm not wrong the 3 main functions that could benefit from NK's methods would be filter_physio(), interpolate_physio() and peakfind_physio().
What you could do if you don't want to have NK as a dependency is to have it as a optional dependency, i.e., add an argument like
method = "peakdet"
(in which case it would do what it currently does), and if method is something else then it tries to load NK (see a usage example here).For the peak detection method, we could outsource it (or duplicate it if you go for optional dependency) in NK either as a new ecg peak detection method or as a more general agnostic peak detection method.