Suppose I want to predict whether a patient has post-operative complications. In addition to some 'usual' regressors, such as age and weights, I also have access to variables that are measured over some time, such as their heartrate during surgery and such. I am wondering how to include such variables that are measured over time in my model. One further complication is that not all surgeries last equally long.
One option that seems quite intuitive is reducing such data to some variables such as the median, standard deviation, frequency of dropping below a certain level, etc., which can then be included as usual in a model. But I was wondering whether there is some better or more systematic approach to do this.
I am also interested in knowing relevant terms that I could google to learn more about this case.