Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Teeth3DS+: An Extended Benchmark for Intraoral 3D Scans Analysis

Achraf Ben-Hamadou, Nour Neifar, Ahmed Rekik, Oussama Smaoui, Firas Bouzguenda, Sergi Pujades, Edmond Boyer, Edouard Ladroit

TL;DR

The paper introduces Teeth3DS+, the first public benchmark for intraoral 3D scan analysis, addressing the shortage of high-quality datasets for teeth segmentation, labeling, and landmark detection. It details a large-scale, GDPR-compliant data collection from 900 patients using multiple IOS devices, paired with an eight-step annotation workflow that includes UV mapping and harmonic parameterization to enable precise per-tooth labeling and boundary propagation. It also extends to dental landmark annotations for a complementary Teeth3DSLand task, with a clear training/testing split used in MICCAI challenges, and provides data in accessible JSON/OBJ formats hosted publicly. Overall, Teeth3DS+ offers a standardized, clinically validated resource to benchmark and accelerate automated CAD tools in dentistry, supporting broader 3D modeling and anomaly detection applications.

Abstract

Intraoral 3D scans analysis is a fundamental aspect of Computer-Aided Dentistry (CAD) systems, playing a crucial role in various dental applications, including teeth segmentation, detection, labeling, and dental landmark identification. Accurate analysis of 3D dental scans is essential for orthodontic and prosthetic treatment planning, as it enables automated processing and reduces the need for manual adjustments by dental professionals. However, developing robust automated tools for these tasks remains a significant challenge due to the limited availability of high-quality public datasets and benchmarks. This article introduces Teeth3DS+, the first comprehensive public benchmark designed to advance the field of intraoral 3D scan analysis. Developed as part of the 3DTeethSeg 2022 and 3DTeethLand 2024 MICCAI challenges, Teeth3DS+ aims to drive research in teeth identification, segmentation, labeling, 3D modeling, and dental landmarks identification. The dataset includes at least 1,800 intraoral scans (containing 23,999 annotated teeth) collected from 900 patients, covering both upper and lower jaws separately. All data have been acquired and validated by experienced orthodontists and dental surgeons with over five years of expertise. Detailed instructions for accessing the dataset are available at https://crns-smartvision.github.io/teeth3ds

Teeth3DS+: An Extended Benchmark for Intraoral 3D Scans Analysis

TL;DR

The paper introduces Teeth3DS+, the first public benchmark for intraoral 3D scan analysis, addressing the shortage of high-quality datasets for teeth segmentation, labeling, and landmark detection. It details a large-scale, GDPR-compliant data collection from 900 patients using multiple IOS devices, paired with an eight-step annotation workflow that includes UV mapping and harmonic parameterization to enable precise per-tooth labeling and boundary propagation. It also extends to dental landmark annotations for a complementary Teeth3DSLand task, with a clear training/testing split used in MICCAI challenges, and provides data in accessible JSON/OBJ formats hosted publicly. Overall, Teeth3DS+ offers a standardized, clinically validated resource to benchmark and accelerate automated CAD tools in dentistry, supporting broader 3D modeling and anomaly detection applications.

Abstract

Intraoral 3D scans analysis is a fundamental aspect of Computer-Aided Dentistry (CAD) systems, playing a crucial role in various dental applications, including teeth segmentation, detection, labeling, and dental landmark identification. Accurate analysis of 3D dental scans is essential for orthodontic and prosthetic treatment planning, as it enables automated processing and reduces the need for manual adjustments by dental professionals. However, developing robust automated tools for these tasks remains a significant challenge due to the limited availability of high-quality public datasets and benchmarks. This article introduces Teeth3DS+, the first comprehensive public benchmark designed to advance the field of intraoral 3D scan analysis. Developed as part of the 3DTeethSeg 2022 and 3DTeethLand 2024 MICCAI challenges, Teeth3DS+ aims to drive research in teeth identification, segmentation, labeling, 3D modeling, and dental landmarks identification. The dataset includes at least 1,800 intraoral scans (containing 23,999 annotated teeth) collected from 900 patients, covering both upper and lower jaws separately. All data have been acquired and validated by experienced orthodontists and dental surgeons with over five years of expertise. Detailed instructions for accessing the dataset are available at https://crns-smartvision.github.io/teeth3ds
Paper Structure (7 sections, 6 figures, 1 table)

This paper contains 7 sections, 6 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methods
  3. Patients and intra-oral scans collection
  4. Data annotation and processing
  5. Teeth detection, segmentation, and labeling
  6. Dental Landmarks
  7. Data records

Figures (6)

  • Figure 1: Illustration of our annotation process. An input 3D IOS is annotated following eight steps, beginning with preprocessing and pose normalization and ending with clinical validation. The clinical validator can return the annotation to steps 2, 4, or 7, depending on the raised issue, which respectively corresponds to missing teeth, teeth border issues, or incorrect teeth instance labeling.
  • Figure 2: FDI World Dental Federation notation.
  • Figure 3: Illustration of landmark points on a sample tooth: Mesial (red), Distal (green), Cusp (blue), Inner (yellow), Outer (cyan), and Facial Axis (magenta) points.
  • Figure 4: Dataset structure for both training and testing phases. All data are structured by jaws and patient ID. Stl files represent the raw intra-oral scans (3D mesh). They are converted to obj files for commodity. Json files provide the annotation corresponding to the obj files.
  • Figure 5: Upper (a) and lower (b) statistics regarding the number of teeth per jaw.
  • ...and 1 more figures