Background/Objectives: The integration of AI into dentistry holds promise for improving diagnostic workflows, particularly in the detection of dental pathologies and pre-radiotherapy screening for head and neck cancer patients. This study aimed to develop and validate an AI model for detecting various dental conditions, with a focus on identifying teeth at risk prior to radiotherapy. Methods: A YOLOv8 model was trained on a dataset of 1628 annotated panoramic radiographs and externally validated on 180 radiographs from multiple centers. The model was designed to detect a variety of dental conditions, including periapical lesions, impacted teeth, root fragments, prosthetic restorations, and orthodontic devices. Results: The model showed strong performance in detecting implants, endodontic treatments, and surgical devices, with precision and recall values exceeding 0.8 for several conditions. However, performance declined during external validation, highlighting the need for improvements in generalizability. Conclusions: YOLOv8 demonstrated robust detection capabilities for several dental conditions, especially in training data. However, further refinement is needed to enhance generalizability in external datasets and improve performance for conditions like periapical lesions and bone loss.

Low‐dose computed tomography (LDCT) is a medical imaging modality designed to minimize ionizing radiation exposure while maintaining the ability to produce detailed cross‐sectional images. It is particularly valuable in scenarios requiring repeated imaging, such as cancer screening, follow‐up examinations or pediatric diagnostics, where reducing radiation dose is critical to patient safety. For example, to reduce noise by half, four times the radiation dose is required in the slice. The goal is to achieve post‐processed LDCT images with comparable quality to those obtained from standard‐dose CT imaging. We start with a brief overview of the CT procedures and their limitations. Then we introduce a novel denoising method based on an asymmetric integration of the ConvNeXt backbone with the U‐Net architecture. This novel approach obtained 2–3 times less noise than the original LDCT, having a 10%–20% increase in performance compared to U‐Net implementation, checked against three metrics MSE, SSIMLoss and combinations of both. The results suggest that: (i) augmenting the images with specific noise, obtained from water phantom CT scan test, while training yields superior results compared to generic noise augmentations; (ii) a larger kernel size better extracts features and (iii) a smaller kernel size was mandatory for feature reconstruction.