AutoDetect-mNP: An Unsupervised Machine Learning Algorithm for Automated Analysis of Transmission Electron Microscope Images of Metal Nanoparticles

Abstract

The synthesis quality of artificial inorganic nanocrystals is most often assessed by transmission electron microscopy (TEM) for which new high-throughput advances have dramatically increased both the quantity and information richness of metal nanoparticle (mNP) characterization. Existing automated data analysis algorithms of TEM mNP images generally adopt a supervised approach, requiring a significant effort in human preparation of labelled data that reduces objectivity, efficiency, and generalizability. We have developed an unsupervised algorithm AutoDetect-mNP for automated analysis of TEM images that objectively extracts morphological information of convex mNPs from TEM images based on their shape attributes, requiring little to no human input in the process. The performance of AutoDetect-mNP is tested on two datasets of bright field TEM images of Au nanoparticles with different shapes, demonstrating that the algorithm is quantitatively reliable, and can thus serve as a generalizable measure of the morphology distributions of any mNP synthesis. The AutoDetect-mNP algorithm will aid in future developments of high-throughput characterization of newly synthesized mNPs, and the future advent of time-resolved TEM studies that can investigate reaction mechanisms of mNP synthesis and reactivity.

Mechanism and performance relevance of nanomorphogenesis in polyamide films revealed by quantitative 3D imaging and machine learning

“…Next, we used principal component analysis (PCA), which requires no prior judgment, to project these eight descriptors into PC space (Fig. 5B) ( 40 ). The variances of the first two PCs are 85.4 and 7.7%, respectively, demonstrating sufficient dimension reduction while capturing the dominant features (Fig.…”

Semi-supervised machine learning workflow for analysis of nanowire morphologies from transmission electron microscopy images

“…Next, we test the generalizability of our machine learning workflow on classification of nanoparticle morphologies. We use TEM images in the AutoDetect-mNP dataset 65 which was originally used for shape analysis of gold metal nanoparticles. Here we use those images for a new task -to classify the morphologies that the assembled nanoparticles adopt regardless of nanoparticle shape (short or long nanorods or triangular prisms).…”

“…For the metal nanoparticle (mNP) dataset, we selected TEM images of nanoparticles from the mNP dataset 65 . The mNP dataset contains TEM images of short or long nanorod assemblies and triangular prism of the TEM images, we categorized the assembled nanoparticles into three categories: dispersed nanoparticles, separate clusters, and percolating cluster, with each morphology containing 100 images.…”

Text-mined dataset of gold nanoparticle synthesis procedures, morphologies, and size entities

“…In a third paradigm of scientific investigation, the volume of data-driven approaches to understanding chemistry and materials synthesis is accelerating. These approaches represent a resourceful complement to established computational methods and raw experimentation, and have been proven successful in applications such as materials discovery [14,15], synthesis protocol querying [16], and the simulation and interpretation of characterization results [17,18]. However, data-driven approaches are limited by the completeness and substance of the data resource(s) used.…”