Skip to content

Millimetre-scale magnetocardiography of living rats with thoracotomy

MetadataDetails
Publication Date2022-08-23
JournalCommunications Physics
AuthorsKeigo Arai, Akihiro Kuwahata, Daisuke Nishitani, Ikuya Fujisaki, Ryoma Matsuki
InstitutionsThe University of Tokyo, Tokyo Institute of Technology
Citations72
AnalysisFull AI Review Included
  • Core Achievement: Demonstrated millimetre-scale Magnetocardiography (MCG) of living rats using an ensemble Nitrogen-Vacancy (NV) quantum sensor in bulk diamond, achieving spatial resolution far superior to conventional sensors.
  • Resolution & Proximity: Achieved a current dipole resolution (Aro) of 5.1 mm for the NV sensor, enabled by an invasive thoracotomy procedure that reduced the standoff distance to 0.6-2.0 mm from the heart surface.
  • Sensor Performance: The system demonstrated a magnetic field sensitivity of 140 pT Hz-1/2 across the rat cardiac signal bandwidth (DC ~200 Hz) at room temperature.
  • Modeling Results: The cardiac signal source was successfully explained by a multiple-current-dipole model, indicating vertically distributed current dipoles pointing from the right atrium base to the left ventricular apex.
  • Current Density Mapping: Spatiotemporal dynamics of the cardiac current density were reconstructed using a stream function method, complementary to the dipole model, revealing current stretched in the left ventricle near the Purkinje fiber.
  • Clinical Relevance: This technique provides a contactless, high-resolution method to investigate intra-cardiac electrodynamics, offering a crucial tool for studying the origin and progression of cardiac arrhythmias (e.g., flutter and fibrillation).
ParameterValueUnitContext
NV Magnetic Field Sensitivity140pT Hz-1/2Across DC ~200 Hz cardiac bandwidth
Shot Noise Limit (Calculated)19pT Hz-1/2Sensitivity is 7x above the shot noise limit
Temporal Resolution (Δt)0.34ms10-90% rise time
ODMR Linewidth (FWHM)2.1MHzFull-width-half-maximum
Magnetic Field Dynamic Range±2.5”TLess than 3% loss in sensitivity
NV Center Concentration3.2 × 1017cm-31.8 ppm concentration
P1 Center Concentration2.6 × 1018cm-315 ppm concentration
Laser Wavelength532nmGreen excitation laser
Laser Incident Power (P0)2.0WTypical incident power
NV Dipole Resolution (Aro)5.1mmMinimum separation for distinguishable dipoles
OPM Dipole Resolution (Aro)15mmOptically Pumped Magnetometer (OPM) comparison
NV Standoff Distance (dNV)7.5 ± 0.5mmHeart center to sensor
Heart Surface Proximity0.6-2.0mmHeart surface to diamond sensor
Total Current Dipole Moment (QNV)1.3 ± 0.5 × 103”A mmFitted using multiple-dipole model
Total Current Flow (INV)2.0 ± 0.3 × 102”ACalculated at R-wave peak
NV Ground State Splitting (D)2.87GHzZero-field splitting
Temperature-Frequency Coefficient-74kHz K-1NV ground state splitting susceptibility
  • Diamond Synthesis: Single-crystal diamond was grown via the High-Pressure High-Temperature (HPHT) method (5.5 GPa, 1300-1350 °C) using a Co-Ti-Cu solvent and a (100) seed crystal.
  • NV Creation: NV centers were created by electron beam irradiation (2.0 MeV, 5 × 1017 electrons cm-2 fluence) followed by post-annealing at 1000 °C for 2 h under vacuum.
  • Optical Readout: A 532 nm green laser (2.0 W) was introduced at ~70° incidence (near Brewster’s angle) to maximize absorption. Red fluorescence (637-800 nm) was collected and detected by a silicon photodiode.
  • Magnetometry Scheme: Frequency-modulated Optically Detected Magnetic Resonance (ODMR) was used, employing triple-tone microwaves to excite all three 14N hyperfine peaks simultaneously for enhanced signal contrast.
  • Noise Mitigation: Low-frequency electronic noise was avoided using lock-in upconverting (fmod = 17-25 kHz). Laser noise was cancelled by subtracting a pick-off laser beam signal, and temperature drift was compensated via a microwave feedback system monitoring double resonance peaks.
  • Sample Preparation: Male Wistar rats were anesthetized, tracheotomized, and thoracotomized. The heart was lifted using nylon thread to achieve millimetre-scale proximity to the diamond sensor.
  • Current Source Estimation: Magnetic field maps were analyzed using two complementary methods:
    1. A multiple-current-dipole model (seven vertically distributed dipoles) fitted via nonlinear least squares regression.
    2. A stream-function method (bfieldtools) to rapidly reconstruct the 2D electric current density distribution.
  • Preclinical Cardiac Research:
    • High-resolution mapping of cardiac current dynamics in small animal models (rats) to study the initiation and maintenance mechanisms of complex arrhythmias (e.g., spiral re-entry).
  • Advanced Quantum Sensing:
    • Development of room-temperature, high-sensitivity magnetic sensors for biological applications, offering a viable alternative to cryogenic SQUIDs where short standoff distance is critical.
  • Medical Device Miniaturization:
    • Future integration of miniaturized NV sensors (via on-chip fabrication) onto catheters or endoscopes, enabling invasive, high-resolution MCG mapping directly within the heart during surgical procedures.
  • Non-Invasive Electrophysiology Tools:
    • Enabling rapid, contactless localization of cardiac current sources, potentially reducing the duration and X-ray exposure associated with traditional catheter ablation mapping.
  • Diamond Material Engineering:
    • Utilization of high-quality, high-density NV ensemble diamonds for commercial quantum sensing platforms requiring high sensitivity (pT Hz-1/2) and spatial resolution (mm-scale).
View Original Abstract

Abstract Magnetocardiography is a contactless imaging modality for electric current propagation in the cardiovascular system. Although conventional sensors provide sufficiently high sensitivity, their spatial resolution is limited to a centimetre-scale, which is inadequate for revealing the intra-cardiac electrodynamics such as rotational waves associated with ventricular arrhythmias. Here, we demonstrate invasive magnetocardiography of living rats at a millimetre-scale using a quantum sensor based on nitrogen-vacancy centres in diamond. The acquired magnetic images indicate that the cardiac signal source is well explained by vertically distributed current dipoles, pointing from the right atrium base via the Purkinje fibre bundle to the left ventricular apex. We also find that this observation is consistent with and complementary to an alternative picture of electric current density distribution calculated with a stream function method. Our technique will enable the study of the origin and progression of various cardiac arrhythmias, including flutter, fibrillation, and tachycardia.