2023/06/20 by Katherine Sytwu, Sytwu, Katherine, Luis Rangel DaCosta +3
Biochemistry, Genetics and Molecular Biology · #Advanced Electron Microscopy Techniques and Applications #Cell Image Analysis Techniques #FOS: Computer and information sciences #FOS: Electrical engineering #FOS: Physical sciences #Image and Video Processing (eess.IV) #Machine Learning (cs.LG) #Materials Science (cond-mat.mtrl-sci) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2306.11853
openalex publication_date 2023/06/20 · openalex created_date 2023/06/24 · openalex updated_date 2026/08/01
Neural networks are promising tools for high-throughput and accurate transmission electron microscopy (TEM) analysis of nanomaterials, but are known to generalize poorly on data that is "out-of-distribution" from their training data. Given the limited set of image features typically seen in high-resolution TEM imaging, it is unclear which images are considered out-of-distribution from others. Here, we investigate how the choice of metadata features in the training dataset influences neural network performance, focusing on the example task of nanoparticle segmentation. We train and validate neural networks across curated, experimentally-collected high-resolution TEM image datasets of nanoparticles under controlled imaging and material parameters, including magnification, dosage, nanoparticle diameter, and nanoparticle material. Overall, we find that our neural networks are not robust across microscope parameters, but do generalize across certain sample parameters. Additionally, data preprocessing heavily influences the generalizability of neural networks trained on nominally similar datasets. Our results highlight the need to understand how dataset features affect deployment of data-driven algorithms.