Any way to remove Oximetry drop outs?

[CopaMundial]

Is there any way to break up a SpO2 file into two sessions to eliminate a finger-out period in the middle?
My sensor was loose so I ended up with a period  from 05:18:52 - 06:45:56 where I have readings that bounce between 127 (which I guess is indicating finger out, and seems to be ignored by Oscar) and 35 (which I'm guessing is incomplete contact, but is still calculated by Oscar).

I would like to keep the data before and after this interruption, but don't want these unrealistically low readings to influence my stats.  Is there a way to do that or should I just delete this whole day? 

A little background unrelated to the specific question, but just in case.  I am using Oscar to track pulse ox only right now as I prepare my case to get my insurance to pay for a new PAP device.  I was never able to tolerate my old REMstar Pro M-series very well, even at a relatively low prescription of 8 cmH20 so I have been non-compliant for a decade or more.  I would like to try to get back on the wagon with a more modern machine but I will need to fight with insurance a bit.

[Crimson Nape]

The "I" (eye) model is the only one that allows for multiple sessions. I don't believe anything exists to modify an existing file. The way I understand its file structure is it has a starting time (other than the D+ model) and every data point is an offset to that time.