NMR Environmental Noise Suppression Method Based on Multi-channel Transformer Reconstruction

doi:10.11938/cjmr20253166

Abstract

Abstract:

Portable nuclear magnetic resonance (NMR) spectrometers are susceptible to external electromagnetic interference (EMI), leading to low signal-to-noise ratio and reduced analytical accuracy. This paper proposes a method based on multiple reference coils and deep learning: multi-channel coils are used to acquire ambient electromagnetic noise in parallel, which is fed into a multi-channel Transformer reconstruction (MCTR) network. The network captures long-range dependencies of ambient electromagnetic signals, predicts and removes the environmental noise in the main receiving coil in real time. Experiments show that this method effectively suppresses noise in both simulated and actual NMR data, improves detection performance, outperforms traditional methods, maintains high signal quality, and has strong robustness. It provides effective support for the application of portable NMR in complex electromagnetic environments, and is expected to promote the development of on-site detection.

Key words: portable NMR, multi-channel parallel reception, reference coils, noise suppression, Transformer

CLC Number:

O482.53

ZHAO Jing, BAO Qingjia, ZHANG Zhi, CHEN Gang, WU Zhaobo, HUANG Zhen, LIU Chaoyang. NMR Environmental Noise Suppression Method Based on Multi-channel Transformer Reconstruction[J]. Chinese Journal of Magnetic Resonance, 2026, 43(2): 115-124.

Figures/Tables 7

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Table 1

Fig. 5

Table 2

References 29

[1]	ELSAYED M, ISAH A, HIBA M, et al. A review on the applications of nuclear magnetic resonance (NMR) in the oil and gas industry: laboratory and field-scale measurements[J]. J Pet Explor Prod Technol, 2022, 12(10): 2747-2784.
[2]	HU Y, CHENG K, HE L, et al. NMR-based methods for protein analysis[J]. Anal Chem, 2021, 93(4): 1866-1879. doi: 10.1021/acs.analchem.0c03830 pmid: 33439619
[3]	SPECIALE I, NOTARO A, GARCIA-VELLO P, et al. Liquid-state NMR spectroscopy for complex carbohydrate structural analysis: A hitchhiker's guide[J]. Carbohydr Polym, 2022, 277: 118885. doi: 10.1016/j.carbpol.2021.118885
[4]	BLÜMICH B. Low-field and benchtop NMR[J]. J Magn Reson, 2019, 306: 27-35. doi: S1090-7807(19)30159-4 pmid: 31311709
[5]	MICHAL C A. Low-cost low-field NMR and MRI: Instrumentation and applications[J]. J Magn Reson, 2020, 319: 106800. doi: 10.1016/j.jmr.2020.106800
[6]	HATZAKIS E. Nuclear magnetic resonance (NMR) spectroscopy in food science: A comprehensive review[J]. Compr Rev Food Sci Saf, 2019, 18(1): 189-220. doi: 10.1111/crf3.2019.18.issue-1
[7]	HOLZGRABE U. NMR spectroscopy in pharmaceutical analysis[M]. Amsterdam: Elsevier, 2017.
[8]	LUCHINAT E, CREMONINI M, BANCI L. Radio signals from live cells: the coming of age of in-cell solution NMR[J]. Chem Rev, 2022, 122(10): 9267-9306. doi: 10.1021/acs.chemrev.1c00790 pmid: 35061391
[9]	HOULT D I, RICHARDS R E. The signal-to-noise ratio of the nuclear magnetic resonance experiment[J]. J Magn Reson, 1976, 24(1): 71-85.
[10]	SONG Y Z, YU H Y, CHEN J Z, et al. Topological optimization method for the magnet structure of magnetic resonance logging sensors[J]. Chinese J Magn Reson, 2025, 42(1): 80-88.
	宋彦佐, 于会媛, 陈敬智, 等. 磁共振测井传感器磁体结构的拓扑优化方法[J]. 波谱学杂志, 2025, 42(1): 80-88. doi: 10.11938/cjmr20243131
[11]	XU X, XU Z. Modeling and analysis of litz wire radio frequency (RF) coil in inside-out NMR well logging sensor[J]. IEEE Trans Geosci Remote Sens, 2023, 61: 1-9.
[12]	WANG F, LIU T W, XU Y J, et al. Design of a miniaturized NMR RF probe with an external locking field channel[J]. Chinese J Magn Reson, 2023, 40(3): 332-340.
	王峰, 刘庭伟, 徐雅洁, 等. 一种带外部锁场通道的小型化核磁共振射频探头设计[J]. 波谱学杂志, 2023, 40(3): 332-340. doi: 10.11938/cjmr20223044
[13]	CHANDRA K, AL-HARTHI S, SUKUMARAN S, et al. NMR-based metabolomics with enhanced sensitivity[J]. RSC Adv, 2021, 11(15): 8694-8700. doi: 10.1039/d1ra01103k pmid: 35423404
[14]	BIEDENBÄNDER T, ALADIN V, SAEIDPOUR S, et al. Dynamic nuclear polarization for sensitivity enhancement in biomolecular solid-state NMR[J]. Chem Rev, 2022, 122(10): 9738-9794. doi: 10.1021/acs.chemrev.1c00776
[15]	LUO Q J, WU L Y, WANG Y T, et al. Optimization analysis and experimental verification of FID NMR coil in polarized ³He Systems[J]. Chinese J Magn Reson, 2024, 41(3): 266-275.
	罗庆金, 吴良勇, 王雨婷, 等. ³He极化系统中FID NMR线圈的优化分析及实验验证[J]. 波谱学杂志, 2024, 41(3): 266-275. doi: 10.11938/cjmr20243099
[16]	SVYATOVA A, KOZINENKO V P, CHUKANOV N V, et al. PHIP hyperpolarized [1-¹³C] pyruvate and [1-¹³C] acetate esters via PH-INEPT polarization transfer monitored by ¹³C NMR and MRI[J]. Sci Rep, 2021, 11(1): 5646. doi: 10.1038/s41598-021-85136-2
[17]	BUENO MORAES T, MONARETTO T, COLNAGO L A. Applications of continuous wave free precession sequences in low-field, time-domain NMR[J]. Appl Sci, 2019, 9(7): 1312. doi: 10.3390/app9071312
[18]	RADIC T. Improving the signal-to-noise ratio of surface NMR data due to the remote reference technique[C]// Near Surface 2006-12th EAGE European Meeting of Environmental and Engineering Geophysics. European Association of Geoscientists & Engineers, 2006: cp-14-00068.
[19]	MUELLER-PETKE M, YARAMANCI U. Improving the signal-to-noise ratio of surface-NMR measurements by reference channel based noise cancellation[C]// Near Surface 2010-16th EAGE European Meeting of Environmental and Engineering Geophysics. European Association of Geoscientists & Engineers, 2010: cp-164-00002.
[20]	TIAN B F, LIN J, DUAN Q M, et al. A magnetic resonance signal noise suppression method based on reference coil and variable step adaptive algorithm[J]. Chin J Geophys, 2012, 55(7): 2462-2472.
	田宝凤, 林君, 段清明, 等. 基于参考线圈和变步长自适应的磁共振信号噪声压制方法[J]. 地球物理学报, 2012, 55(7): 2462-2472.
[21]	李同. 地面磁共振早期信号检测及频域分段噪声消减方法研究[D]. 吉林: 吉林大学, 2015.
[22]	COSTABEL S. Noise reduction strategy for underground mrs based on a three-component remote reference antenna[C]// 7th International Workshop on Magnetic Resonance, Changchun, China. 2018: 1-4.
[23]	ZHANG J, DU G, LIN J, et al. Improving the signal-to-noise ratio of underground nuclear magnetic resonance data based on the nearby reference noise cancellation method[J]. IEEE Access, 2019, 7: 75265-75275. doi: 10.1109/Access.6287639
[24]	REARICK T, CHARVAT G L, ROSEN M S, et al. Noise suppression methods and apparatus patent:US, 9797971[P]. 2017-10-10.
[25]	SRINIVAS S A, CAULEY S F, STOCKMANN J P, et al. External dynamic interference estimation and removal (EDITER) for low field MRI[J]. Magn Reson Med, 2022, 87(2): 614-628. doi: 10.1002/mrm.v87.2
[26]	YANG L, HE W, HE Y, et al. Active EMI suppression system for a 50 mT unshielded portable MRI scanner[J]. IEEE Trans Biomed Eng, 2022, 69(11): 3415-3426. doi: 10.1109/TBME.2022.3170450
[27]	ZHAO Y, XIAO L, LIU Y, et al. Electromagnetic interference elimination via active sensing and deep learning prediction for radiofrequency shielding-free MRI[J]. NMR Biomed, 2024, 37(7): e4956. doi: 10.1002/nbm.v37.7
[28]	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]// Proceedings of the 31st International Conference on Neural Information Processing Systems, 2017: 6000-6010.
[29]	周新龙. 低场核磁共振弛豫信号的精确检测方法及其应用研究[D]. 南京: 东南大学, 2020.

实验	方法	MSE (mean±std)	SNR/dB (mean±std)
(a)	TF	0.041±0.017	34.93±1.12
	EDITER	0.041±0.005	36.50±1.07
	MCTR	0.040±0.003	36.54±0.08
(b)	TF	0.215±0.102	22.47±1.62
	EDITER	0.201±0.098	28.93±0.40
	MCTR	0.095±0.006	30.02±0.56
(c)	TF	2.297±0.236	9.07±1.69
	EDITER	0.326±0.058	25.23±1.35
	MCTR	0.102±0.051	27.06±0.52

实验	方法	重建前SNR/dB (mean±std)	重建后SNR/dB (mean±std)
(a)	TF	22.29±0.17	34.30±1.31
	EDITER		35.52±1.01
	MCTR		48.04±0.74
(b)	TF	14.57±0.24	30.36±1.81
	EDITER		31.12±0.86
	MCTR		43.66±0.52
(c)	TF	16.07±0.13	33.51±0.71
	EDITER		35.83±0.35
	MCTR		46.47±0.21
(d)	TF	18.67±0.15	39.83±1.48
	EDITER		41.71±1.05
	MCTR		44.15±0.58