Molecular Identification from AFM images using the IUPAC Nomenclature and Attribute Multimodal Recurrent Neural Networks

TL;DR

This paper introduces a deep learning approach that uses multimodal recurrent neural networks to identify molecular structures from AFM images, framing it as an image captioning task and achieving high accuracy with a large dataset.

Contribution

It presents a novel deep learning architecture that translates AFM images into IUPAC molecular names, moving beyond traditional classification methods.

Findings

High accuracy in molecular identification from AFM images

Successful application of image captioning techniques to molecular data

Large-scale training with the QUAM-AFM dataset

Abstract

Despite being the main tool to visualize molecules at the atomic scale, AFM with CO-functionalized metal tips is unable to chemically identify the observed molecules. Here we present a strategy to address this challenging task using deep learning techniques. Instead of identifying a finite number of molecules following a traditional classification approach, we define the molecular identification as an image captioning problem. We design an architecture, composed of two multimodal recurrent neural networks, capable of identifying the structure and composition of an unknown molecule using a 3D-AFM image stack as input. The neural network is trained to provide the name of each molecule according to the IUPAC nomenclature rules. To train and test this algorithm we use the novel QUAM-AFM dataset, which contains almost 700,000 molecules and 165 million AFM images. The accuracy of the predictions is remarkable, achieving a high score quantified by the cumulative BLEU 4-gram, a common metric in language recognition studies.