Adaptive step size LMS improves ECG detection during MRI at 1.5 and 3 T
We describe a new real-time filter to reduce artefacts on electrocardiogram (ECG) due to magnetic field gradients during MRI. The proposed filter is a least mean square (LMS) filter able to continuously adapt its step size according to the gradient signal of the ongoing MRI acquisition.
Materials and methods
We implemented this filter and compared it, within two databases (at 1.5 and 3 T) with over 6000 QRS complexes, to five real-time filtering strategies (no filter, low pass filter, standard LMS, and two other filters optimized within the databases: optimized LMS, and optimized Kalman filter).
The energy of the remaining noise was significantly reduced (26 vs. 68%, p < 0.001) with the new filter vs. standard LMS. The detection error of our ventricular complex (QRS) detector was: 11% with our method vs. 25% with raw ECG, 35% with low pass filter, 17% with standard LMS, 12% with optimized Kalman filter, and 11% with optimized LMS filter.
The adaptive step size LMS improves ECG denoising during MRI. QRS detection has the same F1 score with this filter than with filters optimized within the database.
KeywordsMagnetic field gradient Electric artefact Noise reduction
Julien Oster contributed to this paper as scientific advisor. Several authors were employed by Schiller SA when this study was designed and conducted.
Compliance with ethical standards
Conflict of interest
Three authors André Guillou, Sarah Ménétré and Grégory Petitmangin are employed by Schiller Medicals, Wissembourg, France and declare this financial link as a potential conflict of interest because the algorithm presented in the manuscript is actually implemented in two schiller commercial devices. The authors Jean-Marc Sellal, Jacques Felblinger and Laurent Bonnemains declare that they have no conflict of interest.
The study complied with the Declaration of Helsinki regarding medical research on human subjects and was approved by a local ethics committee.
All subjects included in the study gave their informed written consent.
- 6.Oster J, Pietquin O, Abacherli R, Kraemer M, Felblinger J (2009) A specific QRS detector for electrocardiography during MRI: using wavelets and local regularity characterization. 2009 IEEE international conference on acoustics, speech and signal process, IEEE, pp 341–344Google Scholar
- 10.Schmidt M, Krug JW, Rose G (2016) Reducing of gradient induced artifacts on the ECG signal during MRI examinations using Wilcoxon filter. Curr Dir Biomed Eng 2:175–178Google Scholar
- 13.Tse ZTH, Dumoulin CL, Clifford GD, Schweitzer J, Qin L, Oster J, Jerosch-Herold M, Kwong RY, Michaud G, Stevenson WG, Schmidt EJ (2014) A 1.5 T MRI-conditional 12-lead electrocardiogram for MRI and intra-MR intervention: 12-lead ECG for MRI and intra-MR intervention. Magn Reson Med 71:1336–1347CrossRefPubMedPubMedCentralGoogle Scholar
- 15.Haykin SO (2014) Adaptive filter theory, 5th edn. Pearson, LondonGoogle Scholar
- 16.Widrow B, Stearns PN (1985) Adaptive signal processing, 1st edn. Prentice Hall, Upper Saddle RiverGoogle Scholar
- 17.R Development Core Team (2009) R: a language and environment for statistical comput ing. R Foundation for Statistical Computing, ViennaGoogle Scholar
- 20.Pillet N, Ayachi M, Frick V, Berviller H, Felblinger J, Blondé JP (2010) A complete device dedicated to ECG signal measurement with integrated 3D Hall sensor for signal correction. 2010 17th IEEE international conference on electronics, circuits and systems, ICECS, pp 227–230Google Scholar