Artificial Intelligence-powered Advances in Phytomedicine: A Systematic Review on Plant-based Drug Discovery Software

Abstract

Herbal medicines contain complex phytochemicals that offer therapeutic potential, but their long-term adverse effects remain challenging to determine. Traditional herbal medicine preparation methods are often labor-intensive, non-standardized, and face difficulties in identifying active ingredients. By improving medication development, optimization, and discovery procedures, Artificial Intelligence (AI) provides innovative ways to get beyond these restrictions. This systematic review aims to evaluate the role of AI-based software such as LEAFSNAP, Deep-Chem, PlantNet, Autodock, MestReNova, XCMS, ADMETlab, SynergyFinder, KEGG, TCMID, and Reactome in enhancing phytomedicine research by streamlining plant identification, bioactive com-pound discovery, molecular modeling, and pharmacokinetic profiling, thereby advancing drug dis-covery and development processes. A systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA guidelines) to explore the integration of AI-based tools in phytomedicine research and drug discovery. Literature searches were performed in Scopus, PubMed, and Google Scholar databases, focusing on studies published between January 2015 and January 2025. Only English-language articles focusing on AI-based software for drug discovery from plant-derived compounds were reviewed. Studies discussing synthetic drug development were ex-cluded; this, along with non-English language, acted as a literature review limitation. Two reviewers independently screened the titles, abstracts, and full texts of the retrieved articles. Any discrepancies in the screening, study selection, and data extraction processes were resolved through discussion and consensus. Through the effective identification of bioactive chemicals and the simplification of intricate studies, the combination of artificial intelligence and computer tools has completely transformed the process of discovering and developing herbal drugs. Artificial intelligence tools like ADMETlab for pharmacokinetics and toxicity assess-ments, DeepChem for molecular activity modelling, and AlphaFold for protein structure prediction have greatly improved the search for new herbal drugs. While AutoDock improves molecular docking simulations by automating ligand-receptor binding predictions for bioactive chemicals, LeafSnap and PlantNet make plant identification simpler. Drug development workflows are improved by databases such as KEGG and TCMID, which combine pathway analysis with conventional medical knowledge. Research on phytomedicine may reach its full potential if AI techniques are improved and traditional knowledge and contemporary technology are better integrated.