Benchmarking deep learning for automated peak detection on GIWAXS data
Authors:
- Constantin VölterCVConstantin Völter
- Vladimir StarostinVSVladimir Starostin
- Dmitry LapkinDLDmitry Lapkin
- Valentin MunteanuVMValentin Munteanu
- Mikhail RomodinMRMikhail Romodin
- Maik HylinskiMHMaik Hylinski
- Alexander GerlachAGAlexander Gerlach
- Alexander HinderhoferAHAlexander Hinderhofer
- Frank SchreiberFSFrank Schreiber
Open Access
- 28 February 2025
- journal article
- research article
- Published by International Union of Crystallography (IUCr) in Journal of Applied Crystallography
- Vol. 58 (2) , 513-522
Abstract
Recent advancements in X-ray sources and detectors have dramatically increased data generation, leading to a greater demand for automated data processing. This is particularly relevant for real-time grazing-incidence wide-angle X-ray scattering (GIWAXS) experiments which can produce hundreds of thousands of diffraction images in a single day at a synchrotron beamline. Deep learning (DL)-based peak-detection techniques are becoming prominent in this field, but rigorous benchmarking is essential to evaluate their reliability, identify potential problems, explore avenues for improvement and build confidence among researchers for seamless integration into their workflows. However, the systematic evaluation of these techniques has been hampered by the lack of annotated GIWAXS datasets, standardized metrics and baseline models. To address these challenges, we introduce a comprehensive framework comprising an annotated experimental dataset, physics-informed metrics adapted to the GIWAXS geometry and a competitive baseline – a classical, non-DL peak-detection algorithm optimized on our dataset. Furthermore, we apply our framework to benchmark a recent DL solution trained on simulated data and discover its superior performance compared with our baseline. This analysis not only highlights the effectiveness of DL methods for identifying diffraction peaks but also provides insights for further development of these solutions.Keywords
Funding Information
- Bundesministerium für Forschung, Technologie und Raumfahrt (05K19VTA, 05K19VTC, VIPR)
- Deutsche Forschungsgemeinschaft (390727645, 460248799)
This publication has 37 references indexed in Scilit:
- Data reduction for serial crystallography using a robust peak finderJournal of Applied Crystallography, 2021
- FACT and FAIR with Big Data allows objectivity in science: The view of crystallographyStructural Dynamics, 2019
- BraggNet: integrating Bragg peaks using neural networksJournal of Applied Crystallography, 2019
- Synchrotron Big Data ScienceSmall, 2018
- ImageJ2: ImageJ for the next generation of scientific image dataBMC Bioinformatics, 2017
- The science is in the dataIUCrJ, 2017
- In situ dynamic observations of perovskite crystallisation and microstructure evolution intermediated from [PbI6]4− cage nanoparticlesNature Communications, 2017
- Deep learningNature, 2015
- Everything SAXS: small-angle scattering pattern collection and correctionJournal of Physics: Condensed Matter, 2013
- P03, the microfocus and nanofocus X-ray scattering (MiNaXS) beamline of the PETRA III storage ring: the microfocus endstationJournal of Synchrotron Radiation, 2012