Automated Single-Particle Reconstruction of Heterogeneous Inorganic Nanoparticles.
Thomas J A SlaterYi-Chi WangGerard M LetebaJhon QuirozPedro H C CamargoSarah J HaighChristopher S AllenPublished in: Microscopy and microanalysis : the official journal of Microscopy Society of America, Microbeam Analysis Society, Microscopical Society of Canada (2021)
Single-particle reconstruction can be used to perform three-dimensional (3D) imaging of homogeneous populations of nano-sized objects, in particular viruses and proteins. Here, it is demonstrated that it can also be used to obtain 3D reconstructions of heterogeneous populations of inorganic nanoparticles. An automated acquisition scheme in a scanning transmission electron microscope is used to collect images of thousands of nanoparticles. Particle images are subsequently semi-automatically clustered in terms of their properties and separate 3D reconstructions are performed from selected particle image clusters. The result is a 3D dataset that is representative of the full population. The study demonstrates a methodology that allows 3D imaging and analysis of inorganic nanoparticles in a fully automated manner that is truly representative of large particle populations.
Keyphrases
- deep learning
- high resolution
- machine learning
- convolutional neural network
- high throughput
- genetic diversity
- cross sectional
- walled carbon nanotubes
- magnetic resonance imaging
- magnetic resonance
- water soluble
- mass spectrometry
- electron microscopy
- computed tomography
- image quality
- photodynamic therapy
- contrast enhanced