Filtern
Dokumenttyp
- Vortrag (2) (entfernen)
Sprache
- Englisch (2)
Referierte Publikation
- nein (2)
Schlagworte
- Electron Microscopy (2) (entfernen)
Organisationseinheit der BAM
Eingeladener Vortrag
- nein (2)
We present a workflow for obtaining fully trained artificial neural networks that can perform automatic particle segmentations of agglomerated, non-spherical nanoparticles from electron microscopy images “from scratch”, without the need for large training data sets of manually annotated images. This is achieved by using unsupervised learning for most of the training dataset generation, making heavy use of generative adversarial networks and especially unpaired image-to-image translation via cycle-consistent adversarial networks. The whole process only requires about 15 minutes of hands-on time by a user and can typically be finished within less than 12 hours when training on a single graphics card (GPU). After training, SEM image analysis can be carried out by the artificial neural network within seconds, and the segmented images can be used for automatically extracting and calculating various other particle size and shape descriptors.
Due to their unique physico-chemical properties, nanoparticles are well established in research and industrial applications. A reliable characterization of their size, shape, and size distribution is not only mandatory to fully understand and exploit their potential and develop reproducible syntheses, but also to manage environmental and health risks related to their exposure and for regulatory requirements. To validate and standardize methods for the accurate and reliable particle size determination nanoscale reference materials (nanoRMs) are necessary. However, there is only a very small number of nanoRMs for particle size offered by key distributors such as the National Institute of Standards and Technology (NIST) and the Joint Research Centre (JRC) and, moreover, few provide certified values. In addition, these materials are currently restricted to polymers, silica, titanium dioxide, gold and silver, which have a spherical shape except for titania nanorods. To expand this list with other relevant nanomaterials of different shapes and elemental composition, that can be used for more than one sizing technique, we are currently building up a platform of novel nanoRMs relying on iron oxide nanoparticles of different shape, size and surface chemistry. Iron oxide was chosen as a core material because of its relevance for the material and life sciences.