CoChat

A movable unshielded magnetocardiography system

Open access

The DOI matches its Crossref record; 60 of the 62 references it cites were found in a registry or an index; 19 of the 55 sources its claims cite were checked for support.

Abstract

Magnetocardiography (MCG), which uses high-sensitivity magnetometers to record magnetic field signals generated by electrical activity in the heart, is a noninvasive method for evaluating heart diseases such as arrhythmia and ischemia. The MCG measurements usually require the participant keeping still in a magnetically shielded room due to the immovable sensor and noisy external environments. These requirements limit MCG applications, such as exercise MCG tests and long-term MCG observations, which are useful for early detections of heart diseases. Here, we introduce a movable MCG system that can clearly record MCG signals of freely behaving participants in an unshielded environment. On the basis of optically pumped magnetometers with a sensitivity of 140 fT/Hz 1/2 , we successfully demonstrated the resting MCG and the exercise MCG tests. Our method is promising to realize a practical movable multichannel unshielded MCG system that nearly sets no limits to participants and brings another kind of insight into the medical diagnosis of heart disease.

Overview

CoChat Summary

TL;DR

The paper presents a movable, unshielded magnetocardiography (MCG) system using self-oscillating optically pumped magnetometers (OPMs) that records cardiac magnetic signals from freely behaving participants in Earth’s field. It demonstrates resting MCG and exercise MCG (eMCG) measurements, including during high-knee exercise, with gradiometric noise suppression and signal processing.

Key findings

  • Each OPM achieved a sensitivity of ∼140 fTrms/Hz1/2, and the two-sensor gradiometer achieved ∼200 fTrms/Hz1/2 with a 15-cm baseline; measured common-mode noise rejection ratio was about 80 at 1 Hz.
  • In eMCG tests, motion artifacts were up to about ∼30 nT, but gradiometric operation suppressed most common-mode motion artifacts to less than 1 nT; breathing artifacts of about 20 pT with a period of ∼3 s were observed at rest.
  • Resting MCG showed clear QRS and T-waves without signal averaging and matched ECG and PPG; pulse transit time was estimated at about 280 ms.
  • In three healthy male participants aged 22–25, eMCG during exercise showed cardiac period decreasing from about 0.65 s at standing still to 0.52 s in stable exercise (115 beats per minute) and increasing to 0.69 s after rest; T-wave amplitude changed from ∼10 pT to ∼15 pT to 6 pT, and R-wave amplitude from ∼35 pT to ∼40 pT to 33 pT across S, E2, and R2 stages.

Methods

  • Design: a movable unshielded MCG system using two robust self-oscillating cesium OPMs in gradiometric mode, with one sensor near the left chest and a reference sensor ∼15 cm away, mounted in a vest; the system operated in Earth’s field (about 54,500 nT in Beijing, China).
  • Data/sample: resting MCG was recorded with participants sitting in a yard away from buildings; eMCG was measured in three healthy male individuals aged 22–25 during stand still (80 s), high-knee exercise (140 s), and standing rest (80 s). ECG and PPG were recorded synchronously for resting measurements.
  • Analysis: raw signals were filtered with a fourth-order 0.6–45 Hz Butterworth band-pass filter and notch filters, then denoised using empirical mode decomposition with interval thresholding (EMD-IT); eMCG features including RR interval, QRS duration, RT interval, T-wave amplitude, and R-wave amplitude were extracted and compared across stages.

Limitations

  • The paper states that technical issues remain for a wearable multichannel system: suppression of magnetic gradient noise, miniaturization of sensors for multichannel operation, and removal of motion artifacts.
  • The paper states that the first-order gradiometer only suppresses common-mode magnetic noise, and that scalar OPM measurements depend on participant orientation relative to the background field, requiring an auxiliary three-axis fluxgate sensor for signal calibration.
  • Evident from the text: the eMCG demonstration used only three healthy male participants aged 22–25, with no patient population; ECG and PPG failed during exercise due to vibration, so exercise-stage comparisons relied on MCG alone.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1126/sciadv.adg1746

Cites

60 of 60 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1. Movable unshielded MCG system using OPMs. (A) Schematic of the OPM. LP, linear polarizer; QWP, quarter-wave plate; PD, photodiode. (B) The noise floor of the MCG system in the natural open environment. The inset shows the noise floor of the two OPMs and the gradiometer in a noisy outdoor environment created by magnetic coils, indicating the ability of the unshielded MCG system to reduce the common-mode magnetic field noise. (C) The overall arrangement of the practical unshielded MCG system. One OPM sensor is placed right on the left chest of participants and the other OPM sensor is placed far away (~15 cm) from the heart. (D) Measured motion artifacts in an eMCG test. The artifacts caused by the motion of OPMs in gradient geomagnetic fields are greatly suppressed by operating OPMs in the gradiometric mode. (E) Time-frequency spectrogram of MCG signals recorded with the gradiometer by using the short-time Fourier transform (STFT).
Fig. 1. Movable unshielded MCG system using OPMs. (A) Schematic of the OPM. LP, linear polarizer; QWP, quarter-wave plate; PD, photodiode. (B) The noise floor of the MCG system in the natural open environment. The inset shows the noise floor of the two OPMs and the gradiometer in a noisy outdoor environment created by magnetic coils, indicating the ability of the unshielded MCG system to reduce the common-mode magnetic field noise. (C) The overall arrangement of the practical unshielded MCG system. One OPM sensor is placed right on the left chest of participants and the other OPM sensor is placed far away (~15 cm) from the heart. (D) Measured motion artifacts in an eMCG test. The artifacts caused by the motion of OPMs in gradient geomagnetic fields are greatly suppressed by operating OPMs in the gradiometric mode. (E) Time-frequency spectrogram of MCG signals recorded with the gradiometer by using the short-time Fourier transform (STFT).CC BY · PubMed Central
Fig. 2. Resting MCG measurement results. (A) Arrangement of the MCG, ECG, and PPG (photoplethysmography) sensors when recording cardiac signals. (B) Cardiac signals recorded with the MCG, ECG, and PPG sensors. The pulse transit time (PTT) is the time delay between the PPG and MCG (ECG) signal for the same cardiac cycle.
Fig. 2. Resting MCG measurement results. (A) Arrangement of the MCG, ECG, and PPG (photoplethysmography) sensors when recording cardiac signals. (B) Cardiac signals recorded with the MCG, ECG, and PPG sensors. The pulse transit time (PTT) is the time delay between the PPG and MCG (ECG) signal for the same cardiac cycle.CC BY · PubMed Central
Fig. 3. The result of eMCG measurements. (A) MCG signals of the participants under the state of standing still (purple line), doing high-knee exercise (red line), and standing rest (blue line). (B) Cardiac period (RR interval) in different states. At the beginning of exercise, the heart period gradually decreases in the E1 stage in about 20 s and then becomes stable in the E2 stage. After the stop of exercise, the cardiac period gradually increases in the R1 stage in about 20 s and becomes stable in the R2 stage. (B to F) The left diagram of each figure shows the variation of the characteristic indicators (cardiac period, QRS duration, RT interval, T-wave amplitude, and R-wave amplitude) with time and different states. The middle one shows the statistical results of these indicators in each stage, in which each gray point represents an indicator result in a cardiac cycle, the dashed line represents the average, the colored box represents the distribution range of 25 to 75%, and the vertical bar represents the range within 1.5 interquartile range. The right one shows the distribution map of these indicators of the S, E2, and R2 stages. Compared with the S stage and the R2 stage, the participants have a shorter cardiac period, QRS duration and RT duration, and larger amplitude of T-wave and R-wave in the E2 stage of exercise.
Fig. 3. The result of eMCG measurements. (A) MCG signals of the participants under the state of standing still (purple line), doing high-knee exercise (red line), and standing rest (blue line). (B) Cardiac period (RR interval) in different states. At the beginning of exercise, the heart period gradually decreases in the E1 stage in about 20 s and then becomes stable in the E2 stage. After the stop of exercise, the cardiac period gradually increases in the R1 stage in about 20 s and becomes stable in the R2 stage. (B to F) The left diagram of each figure shows the variation of the characteristic indicators (cardiac period, QRS duration, RT interval, T-wave amplitude, and R-wave amplitude) with time and different states. The middle one shows the statistical results of these indicators in each stage, in which each gray point represents an indicator result in a cardiac cycle, the dashed line represents the average, the colored box represents the distribution range of 25 to 75%, and the vertical bar represents the range within 1.5 interquartile range. The right one shows the distribution map of these indicators of the S, E2, and R2 stages. Compared with the S stage and the R2 stage, the participants have a shorter cardiac period, QRS duration and RT duration, and larger amplitude of T-wave and R-wave in the E2 stage of exercise.CC BY · PubMed Central
Fig. 4. Principle of the self-oscillating magnetometer. (A) Schematic of the self-oscillating magnetometer. Broad-spectrum light generated by the cesium lamp is collected and collimated by a lens and sent through an optical band-pass filter to remove the far-detuned light. After passing through these polarization optics (LP and QWP), a circularly polarized light is obtained and used for both optical pumping and probing the atoms confined in the vapor cell. The transmitted light is detected by the photodetector, whose signal is amplified, shifted in phase, and then fed back to drive the magnetic field coil. (B) Principle of single-beam self-oscillating magnetometers. Taking the direction of the total magnetic field B→tot that to be measured as the z axis. Without loss of generality, the atomic magnetic moment M→0 generated by the pump light is assumed to lie in yz plane with an angle of θ with respect to the field B→tot. A radio frequency magnetic field B→1 generated by the magnetic field coil is applied along the light to excite the magnetic resonance. Because of the Larmor precession of the atomic magnetic moment around the total magnetic field, only the longitudinal component M0^(∥) parallel to the total magnetic field is preserved and contributes to the magnetic resonance signal. (C) Schematic of the Larmor frequency estimation. The output signal is divided into many overlapped segments of equal length where the time duration of each segment is termed as the gate time. The length of the overlap between two adjacent segments is termed as overlapping time, which determines the sampling rate of the magnetic field measurement.
Fig. 4. Principle of the self-oscillating magnetometer. (A) Schematic of the self-oscillating magnetometer. Broad-spectrum light generated by the cesium lamp is collected and collimated by a lens and sent through an optical band-pass filter to remove the far-detuned light. After passing through these polarization optics (LP and QWP), a circularly polarized light is obtained and used for both optical pumping and probing the atoms confined in the vapor cell. The transmitted light is detected by the photodetector, whose signal is amplified, shifted in phase, and then fed back to drive the magnetic field coil. (B) Principle of single-beam self-oscillating magnetometers. Taking the direction of the total magnetic field B→tot that to be measured as the z axis. Without loss of generality, the atomic magnetic moment M→0 generated by the pump light is assumed to lie in yz plane with an angle of θ with respect to the field B→tot. A radio frequency magnetic field B→1 generated by the magnetic field coil is applied along the light to excite the magnetic resonance. Because of the Larmor precession of the atomic magnetic moment around the total magnetic field, only the longitudinal component M0^(∥) parallel to the total magnetic field is preserved and contributes to the magnetic resonance signal. (C) Schematic of the Larmor frequency estimation. The output signal is divided into many overlapped segments of equal length where the time duration of each segment is termed as the gate time. The length of the overlap between two adjacent segments is termed as overlapping time, which determines the sampling rate of the magnetic field measurement.CC BY · PubMed Central
Fig. 5. MCG signals in different directions. (A) Superposition of the background field (Earth’s field Be) and the cardiac magnetic field Bc. (B) Schematic of the orientation of participants in respect to the Earth’s field. (C) Measured MCG signal in the north-south (NS) direction and the east-west (EW) direction. (D) Measured cardiac period within the measurement time of 120 s. (E and F) Measured amplitude of R-wave and T-wave. The MCG signal in the EW direction shows a larger amplitude than that in the NS direction.
Fig. 5. MCG signals in different directions. (A) Superposition of the background field (Earth’s field Be) and the cardiac magnetic field Bc. (B) Schematic of the orientation of participants in respect to the Earth’s field. (C) Measured MCG signal in the north-south (NS) direction and the east-west (EW) direction. (D) Measured cardiac period within the measurement time of 120 s. (E and F) Measured amplitude of R-wave and T-wave. The MCG signal in the EW direction shows a larger amplitude than that in the NS direction.CC BY · PubMed Central
Fig. 6. MCG signal denoising based on EMD. (A) The resting MCG signal is decomposed into different intrinsic mode functions (IMFs) based on local characteristics of it in time scale. (B) The resting MCG signal before (blue) and after (red) EMD-based denoising. The reconstructed MCG signal is cleaner and easier to analyze.
Fig. 6. MCG signal denoising based on EMD. (A) The resting MCG signal is decomposed into different intrinsic mode functions (IMFs) based on local characteristics of it in time scale. (B) The resting MCG signal before (blue) and after (red) EMD-based denoising. The reconstructed MCG signal is cleaner and easier to analyze.CC BY · PubMed Central

Verification

Claims the paper makes

9 Valid · 4 Not valid · 10 Not checked · 55 Own results and interpretations

20 more claims

Create a free account to see all 78 claims and every source verdict.

See all claims — free account
  • It may state two separate claims as one. 32
  • Probably not a substantial claim. 24

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Research works printed but not linked: 3

  1. J. P. Wikswo Jr Theoretical aspects of the ECG-MCG relationship in Biomagnetism An Interdisciplinary Approach S. J. Williamson G.-L. Romani L. Kaufman I. Modena Eds. (Plenum 1983) pp. 311–326.
  2. P. Siltanen, Magnetocardiography. Compr. Electrocardiol 2, 1405–1438 (1989).
  3. M. Blanco-Velasco B. Weng K. E. Barner A new ECG enhancement algorithm for stress ECG tests in 2006 Computers in Cardiology Valencia Spain (IEEE 2006) pp. 917–920.

Cited by

See all 116

Metadata

Authors
Xiao, W., Sun, C., Shen, L., Feng, Y., Liu, M., Wu, Y., Liu, X., Wu, T., Peng, X., & Guo, H.
Affiliations
Peking University (CN)
Corresponding author
  • Xiao, Wei
  • Sun, Chenxi
  • Wu, Teng
  • Guo, Hong
Venue
Science Advances
Published
2023-03-31
Volume
9
Issue
13
Pages
eadg1746-eadg1746
ISSN
2375-2548 (ISSN-L)
OpenAlex
W4361302002
Cited by
116
References
62
Citations by year
2026: 37 · 2025: 41 · 2024: 27 · 2023: 11
FWCI
13.44
Type
journal article
Language
en
Open access
Licence
cc-by-nc
Locations
  • Science Advances (publishedVersion, cc-by-nc)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
Funders
National Natural Science Foundation of China (61531003, 91436210, 62071012, 61571018)
Topics
  • Physical Sciences › Physics and Astronomy › Atomic and Molecular Physics, and Optics › Atomic and Subatomic Physics Research
  • Health Sciences › Medicine › Cardiology and Cardiovascular Medicine › ECG Monitoring and Analysis
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Magnetic Field Sensors Techniques
Keywords
magnetocardiography, optically pumped magnetometers, heart disease diagnosis, cardiac arrhythmias, ischemia
MeSH terms
Arrhythmias, Cardiac, Exercise Test, Heart, Heart Diseases, Humans, Magnetocardiography

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.