Scanning probe microscopy in the age of machine learning

Abstract

Scanning probe microscopy (SPM) has revolutionized our ability to explore the nanoscale world, enabling the imaging, manipulation, and characterization of materials at the atomic and molecular level. However, conventional SPM techniques suffer from limitations, such as slow data acquisition, low signal-to-noise ratio, and complex data analysis. In recent years, the field of machine learning (ML) has emerged as a powerful tool for analyzing complex datasets and extracting meaningful patterns and features in multiple fields. The combination of ML with SPM techniques has the potential to overcome many of the limitations of conventional SPM methods and unlock new opportunities for nanoscale research. In this review article, we will provide an overview of the recent developments in ML-based SPM, including its applications in topography imaging, surface characterization, and secondary imaging modes, such as electrical, spectroscopic, and mechanical datasets. We will also discuss the challenges and opportunities of integrating ML with SPM techniques and highlight the potential impact of this interdisciplinary field on various fields of science and engineering.