MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641107212 A) filed by Dayananda Sagar College Of Engineering on September 07, 2026, for System And Method For Automated Three-Dimensional Dental Arch Segmentation And Orthodontic Diagnosis.
Inventors include Prof. Ramya K; Dr. Dharmesh H S; Abhitha A Bhat; Pratiksha; Pruthvi K S; Suvedh V; and Dr. Aruna M G.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: The invention discloses an AI-powered system and method for automated 3D dental arch segmentation, tooth identification, and orthodontic diagnosis, leveraging a deep learning point cloud neural network. The system accepts 3D dental arch scans in NumPy Compressed (NPZ) format, where each point is encoded as (X, Y, Z) spatial coordinates of the scanned dental surface. At its core, a custom PointNetSeg neural network trained on approximately 1,800 arch scans classifies every point in the scan into one of 50 classes, comprising 49 individual tooth IDs and one background/gum class. The diagnostic pipeline identifies conditions such as crowding, spacing, and arch misalignment through PCA-based tooth width estimation and second-degree polynomial (parabolic) arch curve fitting. Based on this analysis, the system generates a structured diagnostic report that classifies the dentition into one of five conditions: Normal Alignment, Mild Crowding, Severe Crowding, Generalized Spacing, or Arch Misalignment. The solution is delivered through an interactive web-based interface featuring 3D visualization, alignment simulation, and JWT-secured role-based access control.
Disclaimer: Curated by HT Syndication.
