Deep Learning Computer Vision for Anomaly Detection in Scanning Transmission Electron Microscopy

Prifti, Enea; Klie, Robert F.; Buban, J. P.

doi:10.1017/s1431927622011291

Cited by 5 publications

(4 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many works have been focused on the application of artificial intelligence and machine learning for the analysis of large and/or complex data. 5,[18][19][20][21][22][23][24] However, the algorithms presented in many of the previous works require either a human labelled training data set, or human-made assumptions about the information contained in the data. 5,[20][21][22] Such requirements on the input of human knowledge and assumptions limit the applicability of these algorithms to datasets collected by automated TEM data acquisitions workflows, which not only are large in size, but also could contain unexpected information.…”

Section: Introductionmentioning

confidence: 99%

An artificial intelligence’s interpretation of complex high-resolution in situ transmission electron microscopy data

Wang,

Yan,

Ondry

et al. 2024

Matter

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

An artificial intelligence’s interpretation of complex high-resolution in situ transmission electron microscopy data

Wang,

Yan,

Ondry

et al. 2024

Matter

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

An artificial intelligence's interpretation of complex high-resolution in situ transmission electron microscopy data

Wang

Yan

Ondry

et al. 2023

Preprint

View full text Add to dashboard Cite

In situ transmission electron microscopy (TEM) has enabled researchers to visualize complicated nano- and atomic-scale processes with sub-Angstrom spatial resolution and millisecond time resolution. These processes are often highly dynamical and can be time-consuming to analyze and interpret. Here, we report how variational autoencoders (VAEs), a deep learning algorithm, can provide an artificial intelligence’s interpretation of high-resolution in situ TEM data by condensing and deconvoluting complicated atomic-scale dynamics into a latent space with reduced dimensionality. In this work, we designed a VAEs model with high latent dimensions capable of deconvoluting information from complex high-resolution TEM data. We demonstrate how VAEs with high latent dimensions trained on atomically resolved TEM images of lead sulfide (PbS) nanocrystals is able to capture movements and perturbations of periodic lattices in both simulated and real in situ TEM data. The VAEs model shows capability of detecting and deconvoluting dynamical nanoscale physical processes, such as the rotation of crystal lattices and intraparticle ripening during the annealing of semiconductor nanocrystals. With the help of the VAEs model, we can identify an in situ observation that can serve as a direct experimental evidence of the existence of intraparticle ripening, a process so far has not been recorded under atomically-resolved in situ TEM for colloidal semiconductor nanocrystals. The VAEs model provides a potent tool for facilitating the analysis and interpretation of complex in situ TEM data as a part of an autonomous experimental workflow. It also shows potential to identify unexpected phenomena during the data analysis.

show abstract

“…Many works have been focused on the application of artificial intelligence and machine learning for the analysis of large and/or complex data. 5,16,[22][23][24][25][26][27][28] However, the algorithms presented in many of the previous works require either a human labeled training data set, or human-made assumptions about the information contained in the data. 5,16,[24][25][26] Such requirements on the input of human knowledge and assumptions limit the applicability of these algorithms to datasets collected by automated TEM data acquisitions workflows, which not only are large in size, but also could contain unexpected information.…”

Section: Introductionmentioning

confidence: 99%

An artificial intelligence's interpretation of complex high-resolution in situ transmission electron microscopy data

Wang

Yan

Ondry

et al. 2023

Preprint

View full text Add to dashboard Cite

In situ transmission electron microscopy (TEM) has enabled researchers to visualize complicated nano- and atomic-scale processes with sub-Angstrom spatial resolution and millisecond time resolution. These processes are often highly dynamical and can be time-consuming to analyze and interpret. Here, we report how variational autoencoders (VAEs), a deep learning algorithm, can provide an artificial intelligence’s interpretation of high-resolution in situ TEM data by condensing and deconvoluting complicated atomic-scale dynamics into a latent space with reduced dimensionality. In this work, we designed a VAEs model with high latent dimensions capable of deconvoluting information from complex high-resolution TEM data. We demonstrate how this model with high latent dimensions trained on atomically resolved TEM images of lead sulfide (PbS) nanocrystals is able to capture movements and perturbations of periodic lattices in both simulated and real in situ TEM data. The VAEs model shows capability of detecting and deconvoluting dynamical nanoscale physical processes, such as the rotation of crystal lattices and intraparticle ripening during the annealing of semiconductor nanocrystals. With the help of the VAEs model, we can identify an in situ observation that can serve as a direct experimental evidence of the existence of intraparticle ripening. The VAEs model provides a potent tool for facilitating the analysis and interpretation of complex in situ TEM data as a part of an autonomous experimental workflow.

show abstract

Deep Learning Computer Vision for Anomaly Detection in Scanning Transmission Electron Microscopy

Cited by 5 publications

References 1 publication

An artificial intelligence’s interpretation of complex high-resolution in situ transmission electron microscopy data

An artificial intelligence’s interpretation of complex high-resolution in situ transmission electron microscopy data

An artificial intelligence's interpretation of complex high-resolution in situ transmission electron microscopy data

An artificial intelligence's interpretation of complex high-resolution in situ transmission electron microscopy data

Contact Info

Product

Resources

About