Uncertainty quantification for reliable automatic multiplet classification in 1H NMR spectra

Fischetti, Giulia; Schmid, Nicolas; Henrici, Andreas; Wilhelm, Dirk; Bruderer, Simon; Heitmann, Bjoern; Scarso, Alessandro; Caldarelli, Guido

doi:10.21256/zhaw-29538

Please use this identifier to cite or link to this item: https://doi.org/10.21256/zhaw-29538

Publication type:	Conference poster
Type of review:	Peer review (abstract)
Title:	Uncertainty quantification for reliable automatic multiplet classification in 1H NMR spectra
Authors:	Fischetti, Giulia Schmid, Nicolas Henrici, Andreas Wilhelm, Dirk Bruderer, Simon Heitmann, Bjoern Scarso, Alessandro Caldarelli, Guido
et. al:	No
DOI:	10.21256/zhaw-29538
Proceedings:	Euromar 2023 Programme & Abstract Book
Editors of the parent work:	Prisner, Thomas
Page(s):	350
Conference details:	European Conference on Magnetic Resonance (EUROMAR), Glasgow, United Kingdom, 9-13 July 2023
Issue Date:	Jul-2023
Publisher / Ed. Institution:	ZHAW Zürcher Hochschule für Angewandte Wissenschaften
Language:	English
Subjects:	NMR spectroscopy; Machine learning; Deep learning; Uncertainty quantification
Subject (DDC):	006: Special computer methods 530: Physics
Abstract:	Proton NMR is the fastest and most straightforward of all NMR experimental designs. Unfortunately, it suffers from lengthy annotation times and does not always have a clear and unbiased interpretation. Introducing an automatic procedure for the analysis of NMR data that can ease the chemical compounds characterization while ensuring consistency of the results across the scientific community is still an open challenge. Recently, we introduced a supervised deep learning model that performs automated classification of signal regions for their coupling pattern. Here we show how including uncertainty quantification in deep learning frameworks applied to NMR serves a dual-purpose of increasing the reliability of the prediction and detecting overlapping multiplets.
URI:	https://digitalcollection.zhaw.ch/handle/11475/29538
Fulltext version:	Published version
License (according to publishing contract):	Licence according to publishing contract
Departement:	School of Engineering
Organisational Unit:	Institute of Applied Mathematics and Physics (IAMP)
Published as part of the ZHAW project:	Maschinelles Lernen für NMR-Spektroskopie
Appears in collections:	Publikationen School of Engineering

Files in This Item:

File	Description	Size	Format
2023_Fischetti-etal_Uncertainty-quantification-multiplet-classification_EUROMAR2023-Poster.pdf		2.9 MB	Adobe PDF	View/Open

Show full item record

Fischetti, G., Schmid, N., Henrici, A., Wilhelm, D., Bruderer, S., Heitmann, B., Scarso, A., & Caldarelli, G. (2023). Uncertainty quantification for reliable automatic multiplet classification in 1H NMR spectra [Conference poster]. In T. Prisner (Ed.), Euromar 2023 Programme & Abstract Book (p. 350). ZHAW Zürcher Hochschule für Angewandte Wissenschaften. https://doi.org/10.21256/zhaw-29538

Fischetti, G. et al. (2023) ‘Uncertainty quantification for reliable automatic multiplet classification in 1H NMR spectra’, in T. Prisner (ed.) Euromar 2023 Programme & Abstract Book. ZHAW Zürcher Hochschule für Angewandte Wissenschaften, p. 350. Available at: https://doi.org/10.21256/zhaw-29538.

G. Fischetti et al., “Uncertainty quantification for reliable automatic multiplet classification in 1H NMR spectra,” in Euromar 2023 Programme & Abstract Book, Jul. 2023, p. 350. doi: 10.21256/zhaw-29538.

FISCHETTI, Giulia, Nicolas SCHMID, Andreas HENRICI, Dirk WILHELM, Simon BRUDERER, Bjoern HEITMANN, Alessandro SCARSO und Guido CALDARELLI, 2023. Uncertainty quantification for reliable automatic multiplet classification in 1H NMR spectra. In: Thomas PRISNER (Hrsg.), Euromar 2023 Programme & Abstract Book. Conference poster. ZHAW Zürcher Hochschule für Angewandte Wissenschaften. Juli 2023. S. 350

Fischetti, Giulia, Nicolas Schmid, Andreas Henrici, Dirk Wilhelm, Simon Bruderer, Bjoern Heitmann, Alessandro Scarso, and Guido Caldarelli. 2023. “Uncertainty Quantification for Reliable Automatic Multiplet Classification in 1H NMR Spectra.” Conference poster. In Euromar 2023 Programme & Abstract Book, edited by Thomas Prisner, 350. ZHAW Zürcher Hochschule für Angewandte Wissenschaften. https://doi.org/10.21256/zhaw-29538.

Fischetti, Giulia, et al. “Uncertainty Quantification for Reliable Automatic Multiplet Classification in 1H NMR Spectra.” Euromar 2023 Programme & Abstract Book, edited by Thomas Prisner, ZHAW Zürcher Hochschule für Angewandte Wissenschaften, 2023, p. 350, https://doi.org/10.21256/zhaw-29538.