search

jcolen / lymphkill

Blood dose and lymphocyte kill simulation
https://aapm.onlinelibrary.wiley.com/journal/24734209
0 stars 0 forks source link

K Wijesooriya -refine the function of regeneration with time #1

Open krishniw opened 5 years ago

krishniw commented 5 years ago

We used a linear fit to simulate the regeneration, but for the overall patient pool, the fractional loss from the beginning of SBRT had a nadir point at day 25 from start of SBRT. overall lymphocyte count after RT = regeneration rate (A) + cell kill rate due to RT that keeps happening months after RT delivery up to certain number of months (B)+ natural cell death (C) A = release of LYA from lymphoid organs (happens only until the LYA recovers to pre Tx value (reduces with time, depends on Pre –tx LYA +dose that kills the LYA) + standard cell generation

Estimate some kill rate (which is a function of absorbed dose and Pre-Tx LYA, since we know both of those are factors), and a regeneration rate curve like what I drew. Then the patient would have stabilized when regeneration is exactly offsetting the kill rate. Making the regeneration and kill rate curves functionals of the initial LYA would eliminate the need for binning entirely. The goal would be to learn the correct functional form.

jcolen commented 5 years ago

Framework for regeneration functions has been written. Need to run regression to find optimal parameters, if they exist. If not, we'll need to think more about the functionals.