Artificial intelligence techniques for prediction of drug synergy in malignant diseases: Past, present, and future

Abstract

In recent years, there has been a surge of interest in the application and acceptance of Artificial Intelligence for challenges involving development, design, and prediction. Artificial Intelligence has not only changed the way we see the world, but it has also offered up new avenues for solving problems. This has been made possible by advances in technology, the availability of large amounts of data generated in various formats, the availability of increasing computational capacity in the form of GPUs and TPUs, and the reduction of costs. The advantages of applying AI in medicine have long been recognized, backed up by ongoing research from numerous institutes, hospitals, and pharmaceutical companies. Drug synergy prediction in malignant diseases is one example of a problem domain that has benefited considerably from AI breakthroughs. Traditionally, finding synergistic drug combinations for malignant diseases by experimental methods has had little success, as promising outcomes may be obtained during trials but may not be achieved during actual treatment due to the development of drug resistance over time. Experimental techniques can only be used for a restricted number of drugs because they are time demanding and expensive. Screening all necessary drug combinations is impractical due to limited resources. The goal of this research is to look at the past, present, and future of AI applications, with an emphasis on drug synergy prediction in malignant diseases using deep learning models. The benefits of utilizing AI to forecast drug synergy are discussed in this paper, as well as future research directions and challenges for applying AI techniques.

Keywords: Deep learning; Drug combinations; Drug synergy prediction; Malignant diseases; Neural networks.