Machine-learning approach for quantified resolvability enhancement of low-dose STEM data

Gambini, Laura; Mullarkey, Tiarnan; Jones, Lewys; Sanvito, Stefano

doi:10.1088/2632-2153/acbb52

Cited by 8 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Current research efforts in the field are focused on the construction of fully automated SPM processes. [316][317][318] Existing AI frameworks possess an algorithmic search of good sample regions and monitor the state of the probe. [319]…”

Section: Scanning Probe Microscopymentioning

confidence: 99%

Expanding the Horizons of Machine Learning in Nanomaterials to Chiral Nanostructures

Kuznetsova,

Coogan,

Botov

et al. 2024

Advanced Materials

View full text Add to dashboard Cite

Machine learning holds significant research potential in the field of nanotechnology, enabling nanomaterial structure and property predictions, facilitating the design of novel materials, and reducing the need for time‐consuming and labour‐intensive experiments and simulations. In contrast to their achiral counterparts, the application of machine learning for chiral nanomaterials and nanostructures is still in its infancy, with a limited number of publications to date. This is despite the great potential of machine learning to advance the development of new sustainable chiral materials with high values of optical activity, circularly polarised luminescence, and enantioselectivity, as well as for the analysis of structural chirality by electron microscopy. In this review, we provide an analysis of machine learning methods used in studying achiral nanomaterials, subsequently offering guidance on adapting and extending this work to chiral nanomaterials. We present an overview of chiral nanomaterials within the framework of synthesis‐structure‐property‐application relationships and provide insights on how to leverage machine learning for the study of these highly complex relationships. We also review and discuss some key recent publications on the application of machine learning for chiral nanomaterials. Finally, the review captures the key achievements, ongoing challenges, and the prospective outlook for this very important research field.This article is protected by copyright. All rights reserved

show abstract

Section: Scanning Probe Microscopymentioning

confidence: 99%

Expanding the Horizons of Machine Learning in Nanomaterials to Chiral Nanostructures

Kuznetsova,

Coogan,

Botov

et al. 2024

Advanced Materials

View full text Add to dashboard Cite

show abstract

“…ML provides robust tools for denoising data sets by effectively distinguishing between noise and essential features. Models such as convolutional neural networks (CNNs) and autoencoders are trained using diverse noisy simulated data , or experimental data. , The trained models adapt and generalize patterns discerned within noisy images, resulting in a significant reduction in image noise while preserving critical details during data postprocessing. This denoising capability is not confined to images alone; it is also applicable to spectroscopy data, , where the background noise can be efficiently suppressed.…”

Section: Application Of Machine Learning and Artificial Intelligence ...mentioning

confidence: 99%

Advances and Opportunities in Closed Gas-Cell Transmission Electron Microscopy

Koo,

Liu,

Cheng

et al. 2024

Chem. Mater.

View full text Add to dashboard Cite

Direct in situ characterizations of the solid–fluid interface on the nanoscale can provide profound implications for addressing bulk-scale enigmas. The advent of closed-cell environmental transmission electron microscopy (E-TEM) enables the implementation of a confined nanoscopic reactor within the high-vacuum microscope. With the encapsulations of reactant fluids and solid species under various stimuli, the whole reaction process can be observed with atomic precision and high temporal resolution. This experimental technique has been adopted widely throughout the field of nanoscience, with applications extending to the synthesis of low-dimensional materials, gas-phase catalysis, and modifications of nanomaterials, where Professor C. N. R. Rao has made substantial contributions over six decades. Here, we delve into the recent representative applications and enhancement strategies of the close gas-cell E-TEM methodology from the early development stages to the latest up-to-date ultrathin (UT) SiN x technique. Remarkable advancements in the capabilities of multimodal data acquisition, including the quantitative electron diffraction, on-site spatiotemporal mapping of the gas molecules, and atomic-resolution real-space imaging in the gas cell E-TEM, are demonstrated. Furthermore, the integration of machine learning (ML)-assisted data acquisition and analysis is anticipated to represent the next major breakthrough, significantly expanding the applicability of gas-cell E-TEM across a wide range of research fields.

show abstract

“…This includes SEM [69] and x-ray tomographic [72] microscopy. DL can be used to remove noise from TEM images, which can improve image quality and facilitate downstream analysis [73][74][75][76][77][78][79][80][81]. DL-based denoising has been applied to SEM [82][83][84] and tomography [85,86] as well.…”

Section: Introductionmentioning

confidence: 99%

Advancing electron microscopy using deep learning

Chen,

Barnard

2024

J. Phys. Mater.

View full text Add to dashboard Cite

Electron microscopy, a sub-field of microanalysis, is a victim of its own success. The widespread use of electron microscopy for imaging molecules and materials has had an enormous impact on our understanding of countless systems and has accelerated impacts in drug discovery and materials design, for electronic, energy, environment and health applications. With this success a bottleneck has emerged, as the rate at which we can collect data has significantly exceeded the rate at which we can analyse it. Fortunately, this has coincided with the rise of advanced computational methods, including data science and machine learning. Deep learning, a sub-field of machine learning capable of learning from large quantities of data such as images, is ideally suited to overcome some of the challenges of electron microscopy at scale. There are a variety of different deep learning approaches relevant to the field, with unique advantages and disadvantages. In this review, we describe some well-established methods, with some recent examples, and introduce some new methods currently emerging in computer science. Our summary of deep learning is designed to guide electron microscopists to choose the right deep learning algorithm for their research and prepare for their digital future.

show abstract

Machine-learning approach for quantified resolvability enhancement of low-dose STEM data

Cited by 8 publications

References 18 publications

Expanding the Horizons of Machine Learning in Nanomaterials to Chiral Nanostructures

Expanding the Horizons of Machine Learning in Nanomaterials to Chiral Nanostructures

Advances and Opportunities in Closed Gas-Cell Transmission Electron Microscopy

Advancing electron microscopy using deep learning

Contact Info

Product

Resources

About