1
|
Schmid N, Bruderer S, Paruzzo F, Fischetti G, Toscano G, Graf D, Fey M, Henrici A, Ziebart V, Heitmann B, Grabner H, Wegner JD, Sigel RKO, Wilhelm D. Deconvolution of 1D NMR spectra: A deep learning-based approach. JOURNAL OF MAGNETIC RESONANCE (SAN DIEGO, CALIF. : 1997) 2023; 347:107357. [PMID: 36563418 DOI: 10.1016/j.jmr.2022.107357] [Citation(s) in RCA: 13] [Impact Index Per Article: 13.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/21/2022] [Revised: 12/01/2022] [Accepted: 12/04/2022] [Indexed: 06/17/2023]
Abstract
The analysis of nuclear magnetic resonance (NMR) spectra to detect peaks and characterize their parameters, often referred to as deconvolution, is a crucial step in the quantification, elucidation, and verification of the structure of molecular systems. However, deconvolution of 1D NMR spectra is a challenge for both experts and machines. We propose a robust, expert-level quality deep learning-based deconvolution algorithm for 1D experimental NMR spectra. The algorithm is based on a neural network trained on synthetic spectra. Our customized pre-processing and labeling of the synthetic spectra enable the estimation of critical peak parameters. Furthermore, the neural network model transfers well to the experimental spectra and demonstrates low fitting errors and sparse peak lists in challenging scenarios such as crowded, high dynamic range, shoulder peak regions as well as broad peaks. We demonstrate in challenging spectra that the proposed algorithm is superior to expert results.
Collapse
Affiliation(s)
- N Schmid
- Zurich University of Applied Sciences (ZHAW), Switzerland; University of Zurich (UZH), Switzerland.
| | | | | | | | | | - D Graf
- Bruker Switzerland AG, Switzerland
| | - M Fey
- Bruker Switzerland AG, Switzerland
| | - A Henrici
- Zurich University of Applied Sciences (ZHAW), Switzerland
| | - V Ziebart
- Zurich University of Applied Sciences (ZHAW), Switzerland
| | | | - H Grabner
- Zurich University of Applied Sciences (ZHAW), Switzerland
| | | | | | - D Wilhelm
- Zurich University of Applied Sciences (ZHAW), Switzerland
| |
Collapse
|
2
|
Anjum MAR, Gonzalez FM, Swain A, Leisen J, Hosseini Z, Singer A, Umpierrez M, Reiter DA. Multi-component T 2 ∗ relaxation modelling in human Achilles tendon: Quantifying chemical shift information in ultra-short echo time imaging. Magn Reson Med 2021; 86:415-428. [PMID: 33590557 DOI: 10.1002/mrm.28686] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/23/2020] [Revised: 12/21/2020] [Accepted: 12/23/2020] [Indexed: 11/10/2022]
Abstract
PURPOSE To examine multi-component relaxation modelling for quantification of on- and off-resonance relaxation signals in multi-echo ultra-short echo time (UTE) data of human Achilles tendon (AT) and compare bias and dispersion errors of model parameters to that of the bi-component model. THEORY AND METHODS Multi-component modelling is demonstrated for quantitative multi-echo UTE analysis of AT and supported using a novel method for determining number of MR-visible off-resonance components, UTE data from six healthy volunteers, and analysis of proton NMR measurements from ex vivo bovine AT. Cramer-Rao lower bound expressions are presented for multi- and bi-component models and parameter estimate variances are compared. Bias error in bi-component estimates is characterized numerically. RESULTS Two off-resonance components were consistently detected in all six volunteers and in bovine AT data. Multi-component model exhibited superior quality of fit, with a marginal increase in estimate variance, when compared to the bi-component model. Bi-component estimates exhibited notable bias particularly in R 2 , 1 ∗ in the presence of off-resonance components. CONCLUSION Multi-component modelling more reliably quantifies tendon matrix water components while also providing quantitation of additional non-water matrix constituents. Further work is needed to interpret the origin of the observed off-resonance signals with preliminary assignments made to chemical groups in lipids and proteoglycans.
Collapse
Affiliation(s)
- Muhammad A R Anjum
- Department of Radiology & Imaging Sciences, School of Medicine, Emory University, Atlanta, Georgia, USA
| | - Felix M Gonzalez
- Department of Radiology & Imaging Sciences, School of Medicine, Emory University, Atlanta, Georgia, USA
| | - Anshuman Swain
- Department of Radiology & Imaging Sciences, School of Medicine, Emory University, Atlanta, Georgia, USA
| | - Johannes Leisen
- School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia, USA
| | - Zahra Hosseini
- MR R&D Collaborations, Siemens Medical Solutions Inc., Atlanta, Georgia, USA
| | - Adam Singer
- Department of Radiology & Imaging Sciences, School of Medicine, Emory University, Atlanta, Georgia, USA
| | - Monica Umpierrez
- Department of Radiology & Imaging Sciences, School of Medicine, Emory University, Atlanta, Georgia, USA
| | - David A Reiter
- Department of Radiology & Imaging Sciences, School of Medicine, Emory University, Atlanta, Georgia, USA.,Department of Orthopedics, School of Medicine, Emory University, Atlanta, Georgia, USA
| |
Collapse
|
3
|
Matviychuk Y, Steimers E, von Harbou E, Holland DJ. Bayesian approach for automated quantitative analysis of benchtop NMR data. JOURNAL OF MAGNETIC RESONANCE (SAN DIEGO, CALIF. : 1997) 2020; 319:106814. [PMID: 32950022 DOI: 10.1016/j.jmr.2020.106814] [Citation(s) in RCA: 13] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 04/29/2020] [Revised: 07/28/2020] [Accepted: 08/24/2020] [Indexed: 06/11/2023]
Abstract
Low-cost, user-friendly benchtop NMR instruments are often touted as a "one-click" solution for data acquisition, however insufficient peak dispersion in their spectra often reduces the accuracy of quantification and requires user expertise with sophisticated processing tools. Our work aims to facilitate the wide acceptance of benchtop NMR instruments as a viable and effective substitute for cryogenic magnets. We propose an algorithmic approach that completely automates the routine analysis of sets of samples with similar compositions - the problem that often underlies many industrial applications concerned with reaction and process monitoring and quality control. Our solution is rooted in the idea of parametric modelling formulated in terms of Bayesian statistics, which effectively incorporates prior knowledge about the studied system (such as concentration-dependent chemical shift changes) that is usually available in industrial applications. Furthermore, the use of quantum mechanical models for chemical species makes our approach invariant to the spectrometer field strength - a necessary prerequisite for the successful analysis of benchtop data. We demonstrate the performance of our method with two representative sets of samples: mixtures of alcohols and acetates, and aqueous mixtures of biologically relevant species. In these examples, our fully automated analysis of benchtop spectra achieves average errors in concentrations of 0.01 mol/mol and 0.02 mol/mol respectively. Our method is competitive with the traditional processing approaches of well resolved high-field data and has the potential to bring the benefits of NMR even to a small chemistry laboratory.
Collapse
Affiliation(s)
- Yevgen Matviychuk
- University of Canterbury, Private Bag 4800, Christchurch 8140, New Zealand.
| | - Ellen Steimers
- Technische Universität Kaiserslautern, Erwin-Schrödinger-Straße 44, 67663 Kaiserslautern, Germany
| | - Erik von Harbou
- Technische Universität Kaiserslautern, Erwin-Schrödinger-Straße 44, 67663 Kaiserslautern, Germany
| | - Daniel J Holland
- University of Canterbury, Private Bag 4800, Christchurch 8140, New Zealand
| |
Collapse
|
4
|
Matviychuk Y, Steimers E, von Harbou E, Holland D. Improving the accuracy of model-based quantitative nuclear magnetic resonance. MAGNETIC RESONANCE (GOTTINGEN, GERMANY) 2020; 1:141-153. [PMID: 37904816 PMCID: PMC10500698 DOI: 10.5194/mr-1-141-2020] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 12/10/2019] [Accepted: 04/30/2020] [Indexed: 11/01/2023]
Abstract
Low spectral resolution and extensive peak overlap are the common challenges that preclude quantitative analysis of nuclear magnetic resonance (NMR) data with the established peak integration method. While numerous model-based approaches overcome these obstacles and enable quantification, they intrinsically rely on rigid assumptions about functional forms for peaks, which are often insufficient to account for all unforeseen imperfections in experimental data. Indeed, even in spectra with well-separated peaks whose integration is possible, model-based methods often achieve suboptimal results, which in turn raises the question of their validity for more challenging datasets. We address this problem with a simple model adjustment procedure, which draws its inspiration directly from the peak integration approach that is almost invariant to lineshape deviations. Specifically, we assume that the number of mixture components along with their ideal spectral responses are known; we then aim to recover all useful signals left in the residual after model fitting and use it to adjust the intensity estimates of modelled peaks. We propose an alternative objective function, which we found particularly effective for correcting imperfect phasing of the data - a critical step in the processing pipeline. Application of our method to the analysis of experimental data shows the accuracy improvement of 20 %-40 % compared to the simple least-squares model fitting.
Collapse
Affiliation(s)
- Yevgen Matviychuk
- Department of Chemical and Process Engineering, University of Canterbury, Private Bag 4800, Christchurch 8140, New Zealand
| | - Ellen Steimers
- Lehrstuhl für Thermodynamik, Technische Universität Kaiserslautern, Erwin-Schrödinger-Straße 44, Kaiserslautern 67663, Germany
| | - Erik von Harbou
- Lehrstuhl für Thermodynamik, Technische Universität Kaiserslautern, Erwin-Schrödinger-Straße 44, Kaiserslautern 67663, Germany
- current address: BASF SE, Research and Development, Ludwigshafen, Germany
| | - Daniel J. Holland
- Department of Chemical and Process Engineering, University of Canterbury, Private Bag 4800, Christchurch 8140, New Zealand
| |
Collapse
|
5
|
Grootveld M, Percival B, Gibson M, Osman Y, Edgar M, Molinari M, Mather ML, Casanova F, Wilson PB. Progress in low-field benchtop NMR spectroscopy in chemical and biochemical analysis. Anal Chim Acta 2019; 1067:11-30. [PMID: 31047142 DOI: 10.1016/j.aca.2019.02.026] [Citation(s) in RCA: 62] [Impact Index Per Article: 12.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/17/2019] [Revised: 02/18/2019] [Accepted: 02/19/2019] [Indexed: 02/07/2023]
Abstract
The employment of spectroscopically-resolved NMR techniques as analytical probes have previously been both prohibitively expensive and logistically challenging in view of the large sizes of high-field facilities. However, with recent advances in the miniaturisation of magnetic resonance technology, low-field, cryogen-free "benchtop" NMR instruments are seeing wider use. Indeed, these miniaturised spectrometers are utilised in areas ranging from food and agricultural analyses, through to human biofluid assays and disease monitoring. Therefore, it is both intrinsically timely and important to highlight current applications of this analytical strategy, and also provide an outlook for the future, where this approach may be applied to a wider range of analytical problems, both qualitatively and quantitatively.
Collapse
Affiliation(s)
- Martin Grootveld
- Chemistry for Health/Bioanalytical Sciences Research Group, Leicester School of Pharmacy, De Montfort University, Leicester, LE1 9BH, UK
| | - Benita Percival
- Chemistry for Health/Bioanalytical Sciences Research Group, Leicester School of Pharmacy, De Montfort University, Leicester, LE1 9BH, UK
| | - Miles Gibson
- Chemistry for Health/Bioanalytical Sciences Research Group, Leicester School of Pharmacy, De Montfort University, Leicester, LE1 9BH, UK
| | - Yasan Osman
- Chemistry for Health/Bioanalytical Sciences Research Group, Leicester School of Pharmacy, De Montfort University, Leicester, LE1 9BH, UK
| | - Mark Edgar
- Department of Chemistry, University of Loughborough, Epinal Way, Loughborough, LE11 3TU, UK
| | - Marco Molinari
- School of Applied Sciences, University of Huddersfield, Queensgate, Huddersfield, HD1 3DH, UK
| | - Melissa L Mather
- Department of Electronic and Electrical Engineering, University of Nottingham, University Park, Nottingham, NG7 2RD, UK
| | | | - Philippe B Wilson
- Chemistry for Health/Bioanalytical Sciences Research Group, Leicester School of Pharmacy, De Montfort University, Leicester, LE1 9BH, UK.
| |
Collapse
|