Due to overlapping clinical features and the critical need for timely diagnosis, accurately distinguishing between dermatological conditions such as Psoriasis and Lichen Planus remains asignificant challenge. This study introduces a novel approach that integrates deep learning with numerical simulations of skin biomechanical properties to improve diagnostic precision. To address the scarcity of clinical data for these conditions, 1000 numerical simulations were conducted using ABAQUS software to generate a robust dataset. Leveraging the ResNet-50 convolutional neural network (CNN), this research incorporates simulation data derived from variations in key biophysical parameters. The final dataset includes 1000 instances equally distributed between Psoriasis and Lichen Planus, with features such as displacement, humidity, age, and sex. To enhance model performance, numerical data were transformed into image-like representations suitable for input into ResNet-50. The model's diagnostic performance was evaluated using 5-fold cross-validation, 3-fold cross-validation, and random splitting. The proposed method achieved a diagnostic accuracy of 99.8% under 5-fold cross-validation, surpassing prior studies and demonstrating the potential of combining artificial intelligence with biomechanical simulations for real-time, accurate classification of skin diseases to support clinicians in diagnostic decision-making.