Berlin 2024 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
CPP: Fachverband Chemische Physik und Polymerphysik
CPP 34: Emerging Topics in Chemical and Polymer Physics, New Instruments and Methods I
CPP 34.3: Talk
Thursday, March 21, 2024, 10:15–10:30, H 0106
Deciphering Electron Paramagnetic Resonance Spectra via Machine Learning — Shengchun Wang1, Shufei Zhang2, JIhu Su5, •Yi Luo3, and Aiwen Lei4 — 1Department of Medicinal Chemistry, University of Michigan, Ann Arbor, MI, USA — 2Shanghai Artificial Intelligence Laboratory, Shanghai, China — 3Institute of Functional Interfaces, Karlsruhe Institute of Technology, Eggenstein-Leopoldshafen, Germany — 4College of Chemistry and Molecular Sciences, the Institute for Advanced Studies, Wuhan University, Wuhan , China — 5Department of Modern Physics, University of Science and Technology of China, Hefei, China
Elucidating the properties of spin species is one of central issues in modern chemistry, material and biology science. For understanding these species, the rapid identification of isotropic and anisotropic Electron Paramagnetic Resonance (EPR) spectra is foundational. Conventional simulation methods, though detailed, often lag in efficiency. Here we present a hybrid approach integrating conventional computational method and Multi-Layer Perceptron (MLP) algorithms, leverages an extensive literature-derived EPR database, ensuring both speed and accuracy in species identification from EPR spectra. Evaluations validate its superior efficacy, attributing much to its robust database integration. This tool offers a promising bridge between academic rigor, computational efficiency, and EPR literature.
Keywords: Machine Learning; Spectrum Identification; Electron Paramagnetic Resonance