gastropy.fit_sine

Contents

gastropy.fit_sine#

gastropy.fit_sine(signal, sfreq, freq=None, freq_init=0.05)[source]#

Fit a sine wave to an EGG signal using least-squares optimisation.

Minimises the sum of squared residuals between signal and a model A·sin(2πft + φ) using L-BFGS-B. If freq is provided, only phase and amplitude are fitted (faster); otherwise frequency, phase, and amplitude are all free parameters.

Parameters:

  • signal (array_like) – Single-channel EGG signal (1D array).

  • sfreq (float) – Sampling frequency in Hz.

  • freq (float or None, optional) – If provided, the frequency is fixed to this value (Hz) and only phase and amplitude are optimised. If None, all three parameters are fitted jointly. Default is None.

  • freq_init (float, optional) – Initial frequency guess (Hz) used when freq=None. Defaults to 0.05 Hz (normogastric centre, ~3 cpm). Set this to a value near the expected dominant frequency when analysing signals outside the normogastric band (e.g. freq_init=0.1 for tachygastric recordings).

Returns:

result – Fitted parameters:

  • freq_hz : float — fitted (or fixed) frequency in Hz.

  • phase_rad : float — fitted phase in radians.

  • amplitude : float — fitted amplitude (may be negative; a negative amplitude is equivalent to a positive amplitude with a phase shift of π).

  • residual : float — sum of squared residuals at solution.

Return type:

dict

References

Dalmaijer, E. S. (2025). electrography v1.1.1. esdalmaijer/electrography

Examples

>>> import numpy as np
>>> from gastropy.signal import fit_sine
>>> t = np.arange(0, 300, 0.1)
>>> sig = 2.5 * np.sin(2 * np.pi * 0.05 * t + 0.3)
>>> result = fit_sine(sig, sfreq=10.0, freq=0.05)
>>> abs(result["amplitude"] - 2.5) < 0.1
True
Contents