Beyond Textual Data: Predicting Drug-Drug Interactions from Molecular Structure Images using Siamese Neural Networks

TL;DR

This paper introduces a novel approach for predicting drug-drug interactions by using molecular structure images as input to a Siamese neural network, moving beyond traditional text-based methods.

Contribution

It is the first to leverage drug structure images with Siamese networks for DDI prediction, offering a new modality for this task.

Findings

Achieved promising accuracy in DDI prediction using image-based data.

Demonstrated the effectiveness of Siamese networks in modeling drug interactions.

Pioneered the use of molecular structure images in DDI prediction.

Abstract

Predicting and discovering drug-drug interactions (DDIs) is an important problem and has been studied extensively both from medical and machine learning point of view. Almost all of the machine learning approaches have focused on text data or textual representation of the structural data of drugs. We present the first work that uses drug structure images as the input and utilizes a Siamese convolutional network architecture to predict DDIs.