Work

MuSe Net : a deep learning framework for trustworthy multiplet segmentation in 1D 1H NMR spectra

Year: 2024

Type: conference-paper

Abstract: One-dimensional 1H NMR spectroscopy stands out as the quickest and simplest among various experimental setups. Unfortunately, it suffers from lengthy annotation times and does not always have an unambiguous interpretation. Finding an automated approach to streamline the characterization of chemical compounds remains an ongoing challenge that has garnered renewed interest with the emergence of deep learning. Here, we present MuSe Net, Multiplet Segmentation Network, a novel supervised probabilistic deep learning framework that can detect and classify splitting patterns with up to four coupling constants. MuSe Net produces a confidence score to both assess classification reliability and detect signals that do not fit into any other phenotype class.

Source: Zürcher Hochschule für Angewandte Wissenschaften digital collection (Zurich University of Applied Sciences)

DOI: 10.21256/zhaw-31444

Authors: Fischetti, Giulia, Schmid, Nicolas, Bruderer, Simon, Henrici, Andreas, Heitmann, Björn, Scarso, Alessandro, Caldarelli, Guido, Wilhelm, Dirk

Language: English

Topic: Spectroscopy and Chemometric Analyses

Subfield: Analytical Chemistry

Field: Chemistry

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W6925933994

FWCI:

Cites:

Cited by:

Related to: