artifact_removal

Patents related to these algorithms have been provisionally filed.

Common Terms

Notation

Included in comments in the code; typical notation with associated definitions.

User Recommendations

Brief Description of MATLAB Functions/Scripts

We group the functions/scripts as follows:

Algorithm 1

Estimates the frequency and phase shifts by solving a least squares problem that minimizes over frequency, phase shifts, and amplitudes by jointly applying harmonic regression and Newton’s descent. Corresponds to Algorithm 1 (artifact removal algorithm) in [1].

Algorithm 2 (Initialization Algorithm for Algorithm 1)

Estimates the frequency and phase shifts by maximizing the energy using Newton’s ascent. Corresponds to Algorithm 2 (initialization algorithm) in [1]. Also includes an option to use simple harmonic regression with the frequency and phase shift estimates from Algorithm 2 to reconstruct/remove the artifact.

Other

References

[1] P. Chen, et al, “Periodic Artifact Removal with Applications to Deep Brain Stimulation,” preprint, 2022.

Note: This paper is available on this Github as Chen_etal_PeriodicArtifactRemovalDBS.pdf.