基于轻量级深度神经网络的核磁共振波谱降噪  

Noise Reduction of Nuclear Magnetic Resonance Spectroscopy Using Lightweight Deep Neural Network

作　　者：詹昊霖 房启元 刘佳伟 史晓琦 陈心语 黄玉清[2] 陈忠[2] Haolin Zhan;Qiyuan Fang;Jiawei Liu;Xiaoqi Shi;Xinyu Chen;Yuqing Huang;Zhong Chen(Department of Biomedical Engineering,Anhui Province Key Laboratory of Measuring Theory and Precision Instrument,School of Instrument Science and Opto-electronics Engineering,Hefei University of Technology,Hefei 230009,China;Department of Electronic Science,Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance,State Key Laboratory of Physical Chemistry of Solid Surfaces,Xiamen University,Xiamen 361005,Fujian Province,China)

机构地区：[1]合肥工业大学仪器科学与光电工程学院生物医学工程系,测量理论与精密仪器安徽省重点实验室,合肥230009 [2]厦门大学电子科学系,福建省等离子体与磁共振重点实验室,固体表面物理化学国家重点实验室,福建厦门361005

出　　处：《物理化学学报》2025年第2期90-97,共8页Acta Physico-Chimica Sinica

基　　金：国家自然科学基金(22204038)资助项目。

摘　　要：核磁共振(NMR)波谱是一种用于探测分子结构和提供定量分析的稳健的非侵入性表征技术。然而,进一步的NMR应用通常受到低灵敏度性能的限制,尤其是对于异核实验。在此,我们提出了一种轻量级的深度学习协议,用于高质量、可靠和快速的NMR波谱降噪。该深度学习(DL)协议具有轻量级的网络优势和快速的计算效率,有效地抑制噪声和伪峰信号,并恢复几乎完全淹没在严重噪声中的目标弱峰,从而实现了可观的信噪比提升。此外,它仅使用物理驱动的仿真NMR数据学习,在频域中实现令人满意的波谱去噪,并允许区分真实信号和噪声伪影。此外,训练的轻量级网络模型通用于一维和多维NMR波谱,并适用于不同的化学样品。因此,本研究呈现的深度学习方法在化学、生物学、材料和生命科学等领域具有应用潜力。Nuclear magnetic resonance(NMR)spectroscopy serves as a robust noninvasive characterization technique for probing molecular structure and providing quantitative analysis,however,further NMR applications are generally confined by the low sensitivity performance,especially for heteronuclear experiments.Herein,we present a lightweight deep learning protocol for high-quality,reliable,and very fast noise reduction of NMR spectroscopy.Along with the lightweight network advantages and fast computational efficiency,this deep learning(DL)protocol effectively reduces noises and spurious signals,and recovers desired weak peaks almost entirely drown in severe noise,thus implementing considerable signal-to-noise ratio(SNR)improvement.Additionally,it enables the satisfactory spectral denoising in the frequency domain and allows one to distinguish real signals and noise artifacts using solely physics-driven synthetic NMR data learning.Besides,the trained lightweight network model is general for one-dimensional and multi-dimensional NMR spectroscopy,and can be exploited on diverse chemical samples.As a result,the deep learning method presented in this study holds potential applications in the fields of chemistry,biology,materials,life sciences,and among others.

关 键 词：核磁共振波谱 人工智能 深度学习 谱图去噪 轻量级网络 

分 类 号：O641[理学—物理化学]