Relate › Nonlinear / curve fitting

Two-phase exponential decay

Two-phase exponential decay fits y = SpanFast·exp(−KFast·x) + SpanSlow·exp(−KSlow·x) + Plateau, capturing biphasic clearance.

What is Two-phase exponential decay?

When the residuals from a one-phase fit show systematic deviation in the early time points, the underlying process has two parallel clearance compartments — fast and slow. SpanFast / SpanSlow are the contributions of each pool; KFast and KSlow are their respective rate constants.

Identifiability requires the two rate constants to be sufficiently different (typical guideline: KFast / KSlow ≥ 5×). Otherwise the model is ill-conditioned and converges to one of many indistinguishable solutions.

When should I use Two-phase exponential decay?

Two-compartment PK models.
Multi-pool radioactive decay.
Any biphasic decay process.

What data does it need?

Numeric time X + signal Y.

What does it report?

SpanFast / SpanSlow / KFast / KSlow / Plateau with CIs.

How do I interpret the result?

Compare one-phase vs two-phase via nonlinear_compare; the F-test tells you whether the extra parameters are justified.