Related Experiment Video
Updated: Apr 14, 2026

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
Published on: October 11, 2018
Enhanced accuracy through machine learning-based simultaneous evaluation: a case study of RBS analysis of multinary
Goele Magchiels1, Niels Claessens2,3, Johan Meersschaut3
1Quantum Solid-State Physics, KU Leuven, Celestijnenlaan 200D, 3001, Leuven, Belgium. goele.magchiels@kuleuven.be.
A new dual-input artificial neural network (ANN) algorithm enhances material analysis by evaluating multiple spectral data sets simultaneously. This approach improves accuracy and precision in complex material characterization, reducing user bias and setup parameter sensitivity.
Area of Science:
- Materials Science
- Computational Science
- Spectroscopy
Background:
- Analyzing large spectral datasets from in situ or in operando experiments requires high accuracy and precision.
- Conventional methods can be susceptible to user bias and inaccuracies in experimental setup parameters.
Purpose of the Study:
- To develop and validate a dual-input artificial neural network (ANN) algorithm for compositional and depth-sensitive analysis of multinary materials.
- To improve the accuracy and precision of spectral data analysis, particularly for complex datasets.
Main Methods:
- A dual-input artificial neural network (ANN) algorithm was developed to simultaneously evaluate spectra from multiple experimental conditions.
- The algorithm was validated using complex Rutherford backscattering spectrometry (RBS) spectra from two scattering geometries.
- Performance was compared against human analysis and single-input ANN analysis.
Main Results:
- The dual-input ANN algorithm provided systematic and precise analysis of complex RBS spectra.
- It demonstrated robustness in handling complex data and minimizing user bias.
- The dual-input ANN showed reduced susceptibility to inaccurately known setup parameters compared to conventional methods.
Conclusions:
- The developed dual-input ANN algorithm offers a robust approach for accurate and precise material characterization.
- This multi-input strategy can be extended to various analytical techniques benefiting from combined measurements under different conditions.
- The method enhances the disentanglement of material property details from complex spectral data.
Related Concept Videos
NMR Spectrometers: Overview
UV–Vis Spectroscopy of Conjugated Systems
One of the factors influencing λmax is...
NMR Spectrometers: Resolution and Error Correction
UV–Vis Spectroscopy: Woodward–Fieser Rules
Mass Spectrometry: Complex Analysis
GC–MS is a powerful hyphenated method commonly used in forensics and environmental...
Atomic Emission Spectroscopy: Overview

