J³SPM AI: An integrated open-source platform for AI-assisted image analysis and image-guided workflows in scanning probe microscopy.

Scanning probe microscopy (SPM), particularly atomic force microscopy (AFM), is essential for characterizing materials at the nanometer scale. However, practical challenges such as image quality degradation and throughput limitations persist, even with advancements in automation. To address these challenges, we developed J³SPM artificial intelligence (AI), an open-source, graphical user interface-based platform designed to facilitate the application of AI to scanning probe microscopy (SPM) workflows. The platform integrates image preprocessing, dataset construction, model training, and inference within a unified environment, enabling AI-assisted analysis without requiring users to develop machine-learning pipelines from scratch. J³SPM AI supports image-based object detection and region-of-interest identification, which can be used to guide zoom-in rescanning and targeted data acquisition. By lowering the technical barrier to AI-based image analysis and workflow integration, J³SPM AI provides a practical framework for incorporating AI-assisted decision-making into conventional and advanced SPM experiments.