Publication

Abstract : The present study develops an Artificial Neural Network (ANN) model to analyze the buoyancy-driven flow of aqueous humor (AH) in the anterior chamber (AC) of the human eye, emphasizing the effects of corneal evaporation and thermal conductivity (TC). These factors significantly influence intraocular fluid and critical parameters for ocular health. The ANN is trained on data generated from simplified Navier–Stokes equations based on lubrication theory, incorporating no-slip boundary conditions at the corneal surface. The model effectively captures how variations in corneal evaporation and TC influence the velocity distribution of AH. The results indicate that increasing the evaporation rate at the corneal surface from 35 W/m 2 to 45 W/m 2 increases the velocity from 1 . 9 × 1 0 − 4 m/s to 2 . 5 × 1 0 − 4 m/s. An increase in TC from 0.47 W/m K to 0.67 W/m K reduces the velocity from 3 . 0 × 1 0 − 4 m/s to 3 . 0 × 1 0 − 4 m/s. Validation against existing experimental and simulation results confirms the predictive accuracy and physical consistency of the ANN.