KinePET¶

KinePET is a TAC-based bounded nonlinear fitter for dynamic PET two-tissue compartment models.

The main interface is the Python package kinepet. A standalone C++ CLI, tcm_fit_tacs, is also available for fitting .npy TAC batches from the command line.

Supported user-facing models:

Model	Meaning
irr	Irreversible two-tissue compartment model
rev	Reversible two-tissue compartment model

KinePET works on flattened time-activity curve batches. The expected TAC layout is (T, N), where T is the number of time points and N is the number of curves.

import kinepet

fit = kinepet.fit_tacs(
    tacs,
    time,
    aif,
    model="rev",
    fit_vb=True,
    threads=8,
)

print(fit.K1.shape)
print(fit.status)

Start with Getting Started for installation and a complete first fit, or go directly to the Python API.