Highly efficient machine learning strategy for low-loss eels characterization: nanophotonic resonances as a case study
Abstract
Spatially-resolved electron energy loss spectroscopy in scanning transmission electron microscopy for materials characterization provides unparalleled access to nanoscale mapping of physical and chemical properties. In particular, the low-loss energy regime contains key details for understanding the optical (e.g., plasmons), electronic (e.g., band gaps), and structural characteristics of materials. However, overlapping resonances, proximity of relevant features to the intense zero-loss peak, and low signal-to-noise ratio complicate the analysis, requiring advanced methods for effective denoising and identification of complex properties. Here, a novel combination of machine learning algorithms is proposed to improve efficiency and interpretation of the low-loss region of EEL spectra and to automatically identify spatially localized information. This work combines uniform manifold approximation and projection for dimensionality reduction with the unsupervised clustering algorithm hierarchical density-based spatial clustering of applications with noise. In a second step, data classified as outliers are reassigned by the supervised support vector machines algorithm. The later supervised stage naturally extends the unsupervised analysis, providing a trained supervised model that classifies new spectrum images based on the previously discovered clusters at negligible computational cost, enabling near real-time deployment on other datasets. This strategy has been applied to silicon/gold nanopillars, used as hybrid metal-semiconductor nanoantennas, to elucidate their nanophotonic resonances (i.e., plasmon, Mie, and hybrid resonances). The procedure enabled accurate mapping of the different resonances in a straightforward manner. The simplicity and versatility of the approach pave the way for studying more complex types of collective excitations and band transitions.
Related
- An evolutionary strategy for highly efficient secretion of multidomain Staphylococcal protein A: N- to C-terminal stability gradient
- An Efficient Enzymatic Strategy for the Diastereoselective Isolation of Tenofovir Alafenamide
- An efficient hypochlorous acid fluorescence imaging strategy based on toehold-mediated strand displacement for distinguishing cancer cells from normal cells
- Quantitative Characterization of RNA-Splinted Ligation as a Multipurpose Post-Synthetic Strategy for High-Density DNA Microarrays Synthesized by Nucleic Acid Photolithography.
- Nanophotonic DyeCycling delivers single-molecule FRET beyond photobleaching to identify heterogenous dynamics in DNA and protein systems
BibTeX
@article{costaledesma2026highly,
title = {{Highly efficient machine learning strategy for low-loss eels characterization: nanophotonic resonances as a case study}},
author = {Vanessa Costa-Ledesma and Daniel Del-Pozo-Bueno and Catalina Coll and Josep Nogués and Borja Sepúlveda and Laura Bocher and Mathieu Kociak and Javier Blanco-Portals and Francesca Peiró and Sónia Estradé},
journal = {npj Computational Materials},
year = {2026},
month = jun,
doi = {10.1038/s41524-026-02171-1},
url = {https://hal.science/hal-05761690v1},
note = {CC BY},
}
RIS
TY - JOUR
TI - Highly efficient machine learning strategy for low-loss eels characterization: nanophotonic resonances as a case study
AU - Vanessa Costa-Ledesma
AU - Daniel Del-Pozo-Bueno
AU - Catalina Coll
AU - Josep Nogués
AU - Borja Sepúlveda
AU - Laura Bocher
AU - Mathieu Kociak
AU - Javier Blanco-Portals
AU - Francesca Peiró
AU - Sónia Estradé
PY - 2026
DA - 2026/06/09
JO - npj Computational Materials
DO - 10.1038/s41524-026-02171-1
UR - https://hal.science/hal-05761690v1
AB - Spatially-resolved electron energy loss spectroscopy in scanning transmission electron microscopy for materials characterization provides unparalleled access to nanoscale mapping of physical and chemical properties. In particular, the low-loss energy regime contains key details for understanding the optical (e.g., plasmons), electronic (e.g., band gaps), and structural characteristics of materials. However, overlapping resonances, proximity of relevant features to the intense zero-loss peak, and low signal-to-noise ratio complicate the analysis, requiring advanced methods for effective denoising and identification of complex properties. Here, a novel combination of machine learning algorithms is proposed to improve efficiency and interpretation of the low-loss region of EEL spectra and to automatically identify spatially localized information. This work combines uniform manifold approximation and projection for dimensionality reduction with the unsupervised clustering algorithm hierarchical density-based spatial clustering of applications with noise. In a second step, data classified as outliers are reassigned by the supervised support vector machines algorithm. The later supervised stage naturally extends the unsupervised analysis, providing a trained supervised model that classifies new spectrum images based on the previously discovered clusters at negligible computational cost, enabling near real-time deployment on other datasets. This strategy has been applied to silicon/gold nanopillars, used as hybrid metal-semiconductor nanoantennas, to elucidate their nanophotonic resonances (i.e., plasmon, Mie, and hybrid resonances). The procedure enabled accurate mapping of the different resonances in a straightforward manner. The simplicity and versatility of the approach pave the way for studying more complex types of collective excitations and band transitions.
SN - 2057-3960
ER -