Visual analysis of drug research and development based on artificial intelligence

Abstract

The iteration of artificial intelligence (AI) technology provides new opportunities for drug research and experimental development. In recent years, AI-based drug research has continuously made new progress and has garnered widespread attention. This study retrieved data from a total of 23,096 papers in AI-based drug research and development from the Web of Science up to May 14, 2024, and conducted bibliometric analysis using VOSviewer software. The results indicated that the AI-based drug research and development is a globally recognized hotspot, and the United States holds a certain authority in this field, while China ranks second in total publication output. The integration of AI technology with drug development primarily involves four stages: drug discovery, preclinical research, clinical trials, and drug manufacturing. So, AI technology has been applied throughout the entire process of drug development. AI-based virtual drug screening and structure-activity relationship analysis started early, while graph neural networks, pre-trained models (Transformer), interpretable AI technology, ChatGPT, and large language models were significantly highlighted in the last 3 years. Moreover, since 2020, AI-based drug repurposing, molecular dynamics simulation, 3D printing, and drug delivery system design have emerged as research hotspots and have been mainly applied to, particularly, on COVID-19, disease prognosis, liver cancer, lung cancer, and immunotherapy.