The prediction of concrete compressive strength using Non-Destructive Testing (NDT) methods, such as the Rebound Hammer Test, has become increasingly popular in the construction industry due to their efficiency and non-invasive nature for Structural Health Monitoring (SHM) However, the accuracy of these traditional methods remains a concern, due to the higher percentage of error in the prediction of strength of concrete structures. This study proposes the application of Artificial Neural Networks (ANN) to enhance the accuracy of concrete strength predictions based on rebound hammer data. Using a dataset of Rebound Hammer Test samples, which is the rebound number, an ANN model was developed, trained, and validated. The results demonstrate a significant improvement, reducing the Mean Absolute Percentage Error (MAPE) to 8.27%. This research highlights the potential of ANNs in improving the reliability of NDT methods and recommends further exploration of artificial intelligence techniques for enhanced prediction accuracy in structural health assessments.