AI-Driven Microscope Paper Published in M&M

How do we operationalize AI for scientific discovery in the electron microscope? In our our article in Microscopy and Microanalysis, we describe the design of a new transmission electron microscope platform combining emerging sparse data analytics with a unique instrument controller. This approach to automated electron microscopy will allow us to examine materials and chemical systems with unprecedented scale, depth, and reliability for important emerging technologies.

From the abstract:

Artificial intelligence (AI) promises to reshape scientific inquiry and enable breakthrough discoveries in areas such as energy storage, quan- tum computing, and biomedicine. Scanning transmission electron microscopy (STEM), a cornerstone of the study of chemical and materials systems, stands to benefit greatly from AI-driven automation. However, present barriers to low-level instrument control, as well as gener- alizable and interpretable feature detection, make truly automated microscopy impractical. Here, we discuss the design of a closed-loop instrument control platform guided by emerging sparse data analytics. We hypothesize that a centralized controller, informed by machine learning combining limited a priori knowledge and task-based discrimination, could drive on-the-fly experimental decision-making. This platform may unlock practical, automated analysis of a variety of material features, enabling new high-throughput and statistical studies.

To view the manuscript, visit: http://doi.org/10.1017/S1431927622012065

To download the article directly, click here.