You are asking if you can take a compound and back-calculate the amount of oxygen present, knowing the ratio of the other elements in the compound. The answer is yes; the issue would be the accuracy of the calculated value with respect to the actual compound.
You can directly measure oxygen using ICP-OES; this is well established in the literature. The issue in your case would be sample introduction.
Ionization Energies: Ar = 15.76eV and Oxygen 13.62eV This is important for the ionization efficiency.
For ICPOES: Oxygen emission spectra: It can be detected in IR region. Between 250-350nm it shows emission lines due to CO molecule and OH. In UV region O peaks below 170nm and gives noise below 190nm.
I would try other techniques rather than ICPs for oxygen determination such as thermal-based techniques and IR determination (thermal desorption gas chromatography, purge and trap, etc.).