Artificial Intelligence-based Prediction of In Vitro Dissolution Profile of Immediate Release Tablets with Near-infrared and Raman Spectroscopy

The objective of the present work was to develop an artificial neural network (ANN) model accurately predict dissolution profile immediate release tablets based on non-destructive spectral data. Six different tablet formulations with varying API (caffeine) and disintegrant (potato starch) concentrations were prepared. near-infrared (NIR) Raman spectra each collected in both reflection transmission modes, then principal component analysis (PCA) conducted. training ANN performed at hidden neuron number from 1 10 order determine optimal neurons layer. best results obtained when a small (1–3) used. In case all four spectroscopic methods, average similarity values (f2) optimized models above 59 for validation tablets, indicating that predicted profiles similar measured curves. exhibited predictive ability. demonstrated potential implementation real-time testing dissolution.