TY - JOUR
T1 - Convolutional Neural Network Analysis of Two-Dimensional Hyperfine Sublevel Correlation Electron Paramagnetic Resonance Spectra
AU - Taguchi, Alexander T.
AU - Evans, Ethan D.
AU - Dikanov, Sergei A.
AU - Griffin, Robert G.
N1 - Funding Information:
This research was supported by grants from the National Institutes of Biomedical Imaging and Bioengineering (EB002804, EB001960, and EB002026) to R.G.G.; a grant from the Chemical Sciences, Geosciences and Biosciences Division of the Office of Basic Energy Sciences at the U.S. Department of Energy (DE-FG02-08ER15960, for pulsed EPR work) to S.A.D.; an NIH F32 Fellowship (GM 123596) to A.T.T.; and an NSF Graduate Research Fellowship (#112237) and the Martin Family Society of Fellows for Sustainability to E.D.E.
Publisher Copyright:
© 2019 American Chemical Society.
PY - 2019/3/7
Y1 - 2019/3/7
N2 - A machine learning approach is presented for analyzing complex two-dimensional hyperfine sublevel correlation electron paramagnetic resonance (HYSCORE EPR) spectra with the proficiency of an expert spectroscopist. The computer vision algorithm requires no training on experimental data; rather, all of the spin physics required to interpret the spectra are learned from simulations alone. This approach is therefore applicable even when insufficient experimental data exist to train the algorithm. The neural network is demonstrated to be capable of utilizing the full information content of two-dimensional 14 N HYSCORE spectra to predict the magnetic coupling parameters and their underlying probability distributions that were previously inaccessible. The predicted hyperfine (a, T) and 14 N quadrupole (K, η) coupling constants deviate from the previous manual analyses of the experimental spectra on average by 0.11 MHz, 0.09 MHz, 0.19 MHz, and 0.09, respectively.
AB - A machine learning approach is presented for analyzing complex two-dimensional hyperfine sublevel correlation electron paramagnetic resonance (HYSCORE EPR) spectra with the proficiency of an expert spectroscopist. The computer vision algorithm requires no training on experimental data; rather, all of the spin physics required to interpret the spectra are learned from simulations alone. This approach is therefore applicable even when insufficient experimental data exist to train the algorithm. The neural network is demonstrated to be capable of utilizing the full information content of two-dimensional 14 N HYSCORE spectra to predict the magnetic coupling parameters and their underlying probability distributions that were previously inaccessible. The predicted hyperfine (a, T) and 14 N quadrupole (K, η) coupling constants deviate from the previous manual analyses of the experimental spectra on average by 0.11 MHz, 0.09 MHz, 0.19 MHz, and 0.09, respectively.
UR - http://www.scopus.com/inward/record.url?scp=85062586123&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85062586123&partnerID=8YFLogxK
U2 - 10.1021/acs.jpclett.8b03797
DO - 10.1021/acs.jpclett.8b03797
M3 - Article
C2 - 30789745
AN - SCOPUS:85062586123
VL - 10
SP - 1115
EP - 1119
JO - Journal of Physical Chemistry Letters
JF - Journal of Physical Chemistry Letters
SN - 1948-7185
IS - 5
ER -