What you are proposing is like trying to calculate the current weather in New York based on cloud cover in Detroit. Sure you can probably find some small statistical correlation but the results won't be useful to a New Yorker trying to plan her day.

If a doctor needs the patient's current platelet count for treatment they're not going to be able to get it through a nasal swab or urine test or whatever. This is restrained by fundamental physical laws. If you're still not getting it then all I can suggest is taking a remedial biology course.