This study focuses on the use of a recurrent neural network (RNN) to This study focused on the use of a recurrent neural network (RNN) to assess pain levels based on heart rate and skin resistance. The performance of the model was investigated across age and gender groups using metrics such as the F-score, Precision, Recall and Accuracy. The results showed significant differences in the model performance depending on these factors, highlighting the need for further calibration and adaptation of the algorithms to individual physiological characteristics. This study contributes to the development of AI-based pain assessment methods and offers new perspectives for future clinical application.