Cargando…
Explainable machine learning for diffraction patterns
Serial crystallography experiments at X-ray free-electron laser facilities produce massive amounts of data but only a fraction of these data are useful for downstream analysis. Thus, it is essential to differentiate between acceptable and unacceptable data, generally known as ‘hit’ and ‘miss’, respe...
Autores principales: | , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
International Union of Crystallography
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10543671/ https://www.ncbi.nlm.nih.gov/pubmed/37791364 http://dx.doi.org/10.1107/S1600576723007446 |
_version_ | 1785114331630796800 |
---|---|
author | Nawaz, Shah Rahmani, Vahid Pennicard, David Setty, Shabarish Pala Ramakantha Klaudel, Barbara Graafsma, Heinz |
author_facet | Nawaz, Shah Rahmani, Vahid Pennicard, David Setty, Shabarish Pala Ramakantha Klaudel, Barbara Graafsma, Heinz |
author_sort | Nawaz, Shah |
collection | PubMed |
description | Serial crystallography experiments at X-ray free-electron laser facilities produce massive amounts of data but only a fraction of these data are useful for downstream analysis. Thus, it is essential to differentiate between acceptable and unacceptable data, generally known as ‘hit’ and ‘miss’, respectively. Image classification methods from artificial intelligence, or more specifically convolutional neural networks (CNNs), classify the data into hit and miss categories in order to achieve data reduction. The quantitative performance established in previous work indicates that CNNs successfully classify serial crystallography data into desired categories [Ke, Brewster, Yu, Ushizima, Yang & Sauter (2018). J. Synchrotron Rad. 25, 655–670], but no qualitative evidence on the internal workings of these networks has been provided. For example, there are no visualization methods that highlight the features contributing to a specific prediction while classifying data in serial crystallography experiments. Therefore, existing deep learning methods, including CNNs classifying serial crystallography data, are like a ‘black box’. To this end, presented here is a qualitative study to unpack the internal workings of CNNs with the aim of visualizing information in the fundamental blocks of a standard network with serial crystallography data. The region(s) or part(s) of an image that mostly contribute to a hit or miss prediction are visualized. |
format | Online Article Text |
id | pubmed-10543671 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2023 |
publisher | International Union of Crystallography |
record_format | MEDLINE/PubMed |
spelling | pubmed-105436712023-10-03 Explainable machine learning for diffraction patterns Nawaz, Shah Rahmani, Vahid Pennicard, David Setty, Shabarish Pala Ramakantha Klaudel, Barbara Graafsma, Heinz J Appl Crystallogr Research Papers Serial crystallography experiments at X-ray free-electron laser facilities produce massive amounts of data but only a fraction of these data are useful for downstream analysis. Thus, it is essential to differentiate between acceptable and unacceptable data, generally known as ‘hit’ and ‘miss’, respectively. Image classification methods from artificial intelligence, or more specifically convolutional neural networks (CNNs), classify the data into hit and miss categories in order to achieve data reduction. The quantitative performance established in previous work indicates that CNNs successfully classify serial crystallography data into desired categories [Ke, Brewster, Yu, Ushizima, Yang & Sauter (2018). J. Synchrotron Rad. 25, 655–670], but no qualitative evidence on the internal workings of these networks has been provided. For example, there are no visualization methods that highlight the features contributing to a specific prediction while classifying data in serial crystallography experiments. Therefore, existing deep learning methods, including CNNs classifying serial crystallography data, are like a ‘black box’. To this end, presented here is a qualitative study to unpack the internal workings of CNNs with the aim of visualizing information in the fundamental blocks of a standard network with serial crystallography data. The region(s) or part(s) of an image that mostly contribute to a hit or miss prediction are visualized. International Union of Crystallography 2023-09-20 /pmc/articles/PMC10543671/ /pubmed/37791364 http://dx.doi.org/10.1107/S1600576723007446 Text en © Shah Nawaz et al. 2023 https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited. |
spellingShingle | Research Papers Nawaz, Shah Rahmani, Vahid Pennicard, David Setty, Shabarish Pala Ramakantha Klaudel, Barbara Graafsma, Heinz Explainable machine learning for diffraction patterns |
title | Explainable machine learning for diffraction patterns |
title_full | Explainable machine learning for diffraction patterns |
title_fullStr | Explainable machine learning for diffraction patterns |
title_full_unstemmed | Explainable machine learning for diffraction patterns |
title_short | Explainable machine learning for diffraction patterns |
title_sort | explainable machine learning for diffraction patterns |
topic | Research Papers |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10543671/ https://www.ncbi.nlm.nih.gov/pubmed/37791364 http://dx.doi.org/10.1107/S1600576723007446 |
work_keys_str_mv | AT nawazshah explainablemachinelearningfordiffractionpatterns AT rahmanivahid explainablemachinelearningfordiffractionpatterns AT pennicarddavid explainablemachinelearningfordiffractionpatterns AT settyshabarishpalaramakantha explainablemachinelearningfordiffractionpatterns AT klaudelbarbara explainablemachinelearningfordiffractionpatterns AT graafsmaheinz explainablemachinelearningfordiffractionpatterns |