Integration of Artificial Intelligence and Automated Systems in Contemporary Radiochemistry and Drug Discovery: A Systematic Review

Abstract

Background: The recent breakthroughs in artificial intelligence (AI) and robotized technology have presented radical changes in radiochemistry and drug discovery with the results of the enhancement of choices accepted according to the data, the efficiency of an experiment and reduction of patterns of development. However, there is scanty comprehensive research done on their combined synergistic impact on these regions.

Objective: In line with the research question, the purpose of the paper and its abstract was to assess the efficacy, performance metrics, and translational capability of AI-driven and automated systems in the contemporary radiochemistry and drug discovery processes.

Methods: A total of 103 records focusing on systematic literature were searched in PubMed/MEDLINE, Scopus, Web of science, and IEEE Xplore and Google Scholar, in accordance with PRISMA 2020. Articles in the field of radiochemistry or drug discovery published between 2014 and 2024 that were exploring AI-based or other automated technologies were considered. Standardized tools were used to extract data and it was quality assessed. They were synthesized with the help of the thematic analysis and narrative synthesis methods to obtain results that pertain to synthesis efficiency, predictive accuracy, speed of development, reproducibility and cost reduction.

Results: A total of one hundred and six peer review articles were found and chosen. Radiochemical synthesis, optimized using AI, showed pooled radiochemical improvements in yield characterized by 32% (23-41) and significant improvements in the time spent in synthesis and increase in reproducibility. Pipelines in drug discovery using AI shortened the mean lead times on hitting and on pre-clinical development by 38 percent and increased efficiency of screening by 100 times. Robotic and automated systems showed better consistency in batch to batch and allowed close loop optimization. The analysis of correlation showed that there are strong positive correlations between the complexity of AI models, the extent of automation, and the general performance results.

Conclusion: The combination of AI-based and automated solutions will be a paradigm shift in radiochemistry and drug discovery, and will provide significant increases in efficiency and accuracy and scalability. The technologies make available closed-loop experimental workflows that are data-driven and have extensive potential implications on radiopharmaceutical production and pharmaceutical innovation. Large-scale clinical and industrialization requires prospective validation and alignment of efforts to establish regulatory harmonization.