# Why use a notch filter to implement a bandpass

I have seen in a few papers, for instance in this one, the following method :

1. Find an initial "frequency of interest" (FOI)
2. Filter the input signal with a Notch filter around the FOI
3. Subtract the result from the initial signal
4. The result is the "interesting part" of the signal, with mainly frequencies around the FOI

Why would one do that rather than applying directly a bandpass filter of similar order and bandwidth ?