Overjet Charting Assist

Overjet Charting Assist is an AI-powered software designed to aid dental professionals by analyzing 2D dental radiographs (bitewing, periapical, and panoramic) to detect and segment tooth anatomy, assign tooth numbers, and identify existing restorative treatments. Its methodology relies on machine learning algorithms processing image data to provide semi-automated charting suggestions, with clinical studies demonstrating robust performance in identifying dental structures and a significant reduction in manual charting time, thereby supporting clinical workflow efficiency.

Overjet Charting Assist is a medical image management and processing system designed to aid dental professionals in the creation of dental charts. The system utilizes artificial intelligence-based image analysis to interpret 2D dental radiographs, including bitewing (BW), periapical (PA), and panoramic (Pano) images. Its core function is to detect and segment natural dental structures, such as tooth anatomy (enamel, pulp) and tooth numbering, as well as identify dental structures resulting from past restorative treatments, including implants, crowns, endodontic treatments, and fillings. The device is intended for use with BW and PA images for patients aged 5 and older with primary and/or permanent teeth, and with Pano radiographs for patients with only permanent teeth, operating as a cloud-based platform.\n\nStandalone performance testing on 634 diverse radiographic images demonstrated the device's capabilities. For past restorative structures, the system achieved an overall tooth-level sensitivity of 88.3% and high specificities for individual restoration types (e.g., 0.986 for fillings, 0.994 for crowns). Dental tooth anatomy (enamel/pulp) was detected with an overall sensitivity of 95.9%, with specific sensitivities of 94.9% for enamel and 85.6% for pulp. Tooth numbering classification accuracy reached 98.8%. The system significantly reduced manual charting operations by an average of 80.5%, enhancing efficiency. Performance was consistently validated across various demographics, U.S. geographical regions, and multiple X-ray sensor manufacturers, meeting predefined acceptance criteria; however, the device is intended as an assistive tool and does not replace a clinician's comprehensive review.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the AI algorithm differentiate between various types of restorative materials (e.g., amalgam vs. composite fillings) and how does this affect detection accuracy?; What is the impact of image quality variations (e.g., under/overexposure, artifacts, sensor types beyond those listed) on the performance of Overjet Charting Assist, particularly for subtle features?; The NPV for dental tooth anatomy is 0.710; what are the primary reasons for false negatives in anatomical detection (enamel, pulp), and what steps are being taken to improve this?; How does the 'semi-automatic' nature of the output, allowing user edits, impact overall workflow efficiency and inter-operator variability in final chart accuracy compared to fully manual or fully automatic systems?; For pediatric patients (age 5+), how does the system account for the dynamic nature of mixed dentition (e.g., exfoliating primary teeth, erupting permanent teeth) in tooth numbering and anatomical assessment accuracy?