Multi-Model AI-Assisted Early Detection of Oral Cancer Integrated With Phytochemical Profiling, Antioxidant , And In-Silico Target Evaluation of Boerhavia diffusa Linn

Background: Boerhavia diffusa is traditionally recognized for its pharmacological properties including antioxidant, anti poliferative, anti- inflammatory and antimicrobial activities. Integrating AI and in silico evaluations of plant bioactivities may provide a comprehensive approach to oral cancer. Aim and Objectives: The present study is designed to investigate the therapeutic potential for B. diffusa plant extracts in AI and in silico docking in oral cancer. Materials and Methods: Leaf material of Boerhavia diffusa was collected, shade-dried, and subjected to the ethanolic and aqueous extracts were prepared using Soxhlet apparatus. Phytochemical screening, antioxidant activity DPPH assay. In-silico docking was performed using identified phytoconstituents against apoptotic protein targets retrieved from the RCSB Protein Data Bank including Bcl-2 with ligands from pubchem. An AI model based on CNN was trained and validated for automated oral cancer image classification. Results: The plant extract demonstrated using preliminary tests, antioxidant activity was higher in B.diffusa ethanolic extract compared with aqueous extract and favorable binding affinities in molecular docking. The AI model achieved high accuracy in distinguishing cancerous and non-cancerous images. Conclusion: The integrated approach combining AI-assisted detection with experimental bioactivity and in-silico evaluation of Boerhavia diffusa suggests a multi-disciplinary strategy for improved oral cancer diagnosis and therapeutic exploration.