The document describes an object detection model using region-based convolutional neural networks to detect dental caries and root canals in dental x-ray images. The model was trained on a dataset of dental x-rays labeled with caries and root canal objects. Feature extraction was performed using multiple convolutional layers and concatenation. Anchors, multi-scale training, and hard negative mining were used to improve performance. The model achieved 83.45% accuracy for object detection, outperforming other methods. Future work could include working with larger datasets and real-time detection from video.