ARTIFICIAL INTELLIGENCE IN MEDICINAL PLANT-BASED DRUG DISCOVERY: CURRENT APPLICATIONS, CHALLENGES, AND FUTURE PERSPECTIVES

Abstract

Medicinal plants have played a vital role in supplying therapeutic compounds, but discovering compounds from them needs time, expensive and requires manual intervention. Due to the emergence of Artificial Intelligence, large biological and chemical datasets are being analysed more efficiently and to assist faster identification of promising plant-derived bioactive compounds. This study focuses on how artificial intelligence can be employed in medicinal plant research and drug discovery, and to emphasize how it improves species identification, phytochemical analysis, and to predict bioactive compounds compared with traditional approaches. The study uses a review-based approach, utilizing existing literature, databases, and AI tools employed in medicinal plant research. It discusses machine learning, deep learning, natural language processing, computer vision, molecular docking, virtual screening, and database resources such as DrugBank, PubChem, HERB, IMPPAT, and TKDL. The review shows how AI methods perform strongly in medicinal plant classification, drug–target affinity prediction, synergy prediction, and extraction optimization. Deep Learning models typically exceeded standard machine learning approaches, and AI-based pipelines diminished the need for large scale manual screening while uplifting prediction accuracy. Artificial intelligence is revolutionizing medicinal plant research by making drug discovery quicker, more effective and more systematic. Even though challenges persist in case of data sparsity, interpretability, and ethical concerns. AI has intense potential to support future botanical drug discovery and natural product research.