Research progress of artificial intelligence in high-throughput drug screening

High-throughput screening (HTS) is widely used in modern drug discovery. It enables batch activity testing of compounds and provides important support for the identification of active compounds. However, its screening efficiency and accuracy need to be improved. To address this issue, artificial intelligence (AI) has been gradually integrated into the HTS workflow. Leveraging the advantages of machine learning (ML) and deep learning (DL), AI optimizes applications in structure-based and ligand-based virtual screening, combination drug screening, image analysis, and post-screening data analysis and interpretation, driving the intelligent development of drug discovery. This paper reviews recent research progress in the application of AI in HTS, discusses the implementation of machine learning models, and summarizes key AI applications in HTS-related compound screening, image recognition, and hit identification from complex screening data, aiming to accelerate the development of innovative drugs.