Detection of electrically bridged EEG electrodes via the Electrical Distance metric. More...


Go to the source code of this file.
Classes | |
| struct | UTILSLIB::BridgedElectrodeParams |
Namespaces | |
| namespace | FIFFLIB |
| FIFF file I/O, in-memory data structures and high-level readers/writers. | |
| namespace | UTILSLIB |
| Shared utilities (I/O helpers, spectral analysis, layout management, warp algorithms). | |
Functions | |
| DSPSHARED_EXPORT Eigen::MatrixXd | UTILSLIB::computeElectricalDistance (const Eigen::MatrixXd &data, const FIFFLIB::FiffInfo &info) |
| Compute electrical distance matrix between EEG channels. | |
| DSPSHARED_EXPORT QList< QPair< int, int > > | UTILSLIB::computeBridgedElectrodes (const Eigen::MatrixXd &data, const FIFFLIB::FiffInfo &info, const BridgedElectrodeParams ¶ms=BridgedElectrodeParams()) |
| Detect bridged EEG electrodes. | |
Detection of electrically bridged EEG electrodes via the Electrical Distance metric.
SPDX-License-Identifier: BSD-3-Clause Copyright (c) 2026 MNE-CPP Authors
Two EEG electrodes are said to be bridged when conductive paste, sweat or skin abrasion creates a low-impedance shortcut between them; the recorded signals then become near-identical and any subsequent source localisation is biased. Following Tenke & Kayser (2001, Clin. Neurophysiol. 112) the detector computes the Electrical Distance — the time-domain variance of the difference signal — for every channel pair, and flags pairs whose normalised distance falls below a user-defined threshold (default 0.3).
The implementation operates on the raw broadband signal; no rereferencing or filtering is required up-front. Output is a list of (i, j) channel- index pairs that the caller can mark bad, interpolate, or pass into the SSS bad-channel set.
Definition in file bridged_electrodes.h.