Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Deep Learning in Dental Image Analysis: A Systematic Review of Datasets, Methodologies, and Emerging Challenges

Zhenhuan Zhou, Jingbo Zhu, Yuchen Zhang, Xiaohang Guan, Peng Wang, Tao Li

TL;DR

The paper addresses the need for scalable, accurate dental image analysis by surveying deep learning progress, with a focus on publicly accessible datasets and model architectures. It compiles 260 studies, categorizes data modalities (PAN, PR, CBCT, IOS, IP), and reviews dense-prediction, classification, and other DIA tasks, highlighting CNN-, Transformer-, and hybrid-based methods. Key contributions include a detailed taxonomy of datasets and models, a comprehensive summary of training/evaluation practices, and a candid discussion of public-data limitations and future directions. The work aims to guide researchers toward larger, multimodal, and more generalizable DIA solutions that can robustly support clinical decision-making and dental care delivery.

Abstract

Efficient analysis and processing of dental images are crucial for dentists to achieve accurate diagnosis and optimal treatment planning. However, dental imaging inherently poses several challenges, such as low contrast, metallic artifacts, and variations in projection angles. Combined with the subjectivity arising from differences in clinicians' expertise, manual interpretation often proves time-consuming and prone to inconsistency. Artificial intelligence (AI)-based automated dental image analysis (DIA) offers a promising solution to these issues and has become an integral part of computer-aided dental diagnosis and treatment. Among various AI technologies, deep learning (DL) stands out as the most widely applied and influential approach due to its superior feature extraction and representation capabilities. To comprehensively summarize recent progress in this field, we focus on the two fundamental aspects of DL research-datasets and models. In this paper, we systematically review 260 studies on DL applications in DIA, including 49 papers on publicly available dental datasets and 211 papers on DL-based algorithms. We first introduce the basic concepts of dental imaging and summarize the characteristics and acquisition methods of existing datasets. Then, we present the foundational techniques of DL and categorize relevant models and algorithms according to different DIA tasks, analyzing their network architectures, optimization strategies, training methods, and performance. Furthermore, we summarize commonly used training and evaluation metrics in the DIA domain. Finally, we discuss the current challenges of existing research and outline potential future directions. We hope that this work provides a valuable and systematic reference for researchers in this field. All supplementary materials and detailed comparison tables will be made publicly available on GitHub.

Deep Learning in Dental Image Analysis: A Systematic Review of Datasets, Methodologies, and Emerging Challenges

TL;DR

The paper addresses the need for scalable, accurate dental image analysis by surveying deep learning progress, with a focus on publicly accessible datasets and model architectures. It compiles 260 studies, categorizes data modalities (PAN, PR, CBCT, IOS, IP), and reviews dense-prediction, classification, and other DIA tasks, highlighting CNN-, Transformer-, and hybrid-based methods. Key contributions include a detailed taxonomy of datasets and models, a comprehensive summary of training/evaluation practices, and a candid discussion of public-data limitations and future directions. The work aims to guide researchers toward larger, multimodal, and more generalizable DIA solutions that can robustly support clinical decision-making and dental care delivery.

Abstract

Efficient analysis and processing of dental images are crucial for dentists to achieve accurate diagnosis and optimal treatment planning. However, dental imaging inherently poses several challenges, such as low contrast, metallic artifacts, and variations in projection angles. Combined with the subjectivity arising from differences in clinicians' expertise, manual interpretation often proves time-consuming and prone to inconsistency. Artificial intelligence (AI)-based automated dental image analysis (DIA) offers a promising solution to these issues and has become an integral part of computer-aided dental diagnosis and treatment. Among various AI technologies, deep learning (DL) stands out as the most widely applied and influential approach due to its superior feature extraction and representation capabilities. To comprehensively summarize recent progress in this field, we focus on the two fundamental aspects of DL research-datasets and models. In this paper, we systematically review 260 studies on DL applications in DIA, including 49 papers on publicly available dental datasets and 211 papers on DL-based algorithms. We first introduce the basic concepts of dental imaging and summarize the characteristics and acquisition methods of existing datasets. Then, we present the foundational techniques of DL and categorize relevant models and algorithms according to different DIA tasks, analyzing their network architectures, optimization strategies, training methods, and performance. Furthermore, we summarize commonly used training and evaluation metrics in the DIA domain. Finally, we discuss the current challenges of existing research and outline potential future directions. We hope that this work provides a valuable and systematic reference for researchers in this field. All supplementary materials and detailed comparison tables will be made publicly available on GitHub.
Paper Structure (53 sections, 21 equations, 24 figures, 12 tables)

This paper contains 53 sections, 21 equations, 24 figures, 12 tables.

Table of Contents

  1. Introduction
  2. Dental Imaging and Public Datasets
  3. Dental Imaging
  4. 2D X-ray imaging
  5. Cone-Beam Computed Tomography (CBCT)
  6. Intraoral Scanning (IOS)
  7. Intraoral Photography (IP)
  8. Dental Public Datasets
  9. Basic technologies
  10. Overview
  11. Fundamental Techniques
  12. CNN
  13. Transformer
  14. Hybrid Model
  15. Dental Tasks and Models
  16. ...and 38 more sections

Figures (24)

  • Figure 1: Statistical results from the papers reviewed in this study are visualized in two charts. The bar chart on the left depicts the annual distribution of DL-based, DIA-related papers and public datasets, with the gray line indicating their cumulative total. The pie chart on the right illustrates the regional distribution, highlighting the top three countries with the highest number of studies in each continent.
  • Figure 2: Taxonomy of Deep Learning Applications in Dental Image Analysis. The diagram is divided into two main sections: Dental Public Datasets and Deep Learning Applications in Dental Image Analysis.
  • Figure 3: Dental imaging devices utilized in actual clinical practice, featuring images captured at the Department of Stomatology, First Affiliated Hospital of Nankai University. (a) 2D panoramic radiography device; (b) periapical radiography device; (c) DSLR camera for intraoral photography; (d) radiologist taking intraoral photographs using a camera; (e) radiologist taking intraoral photographs using a mobile device.
  • Figure 4: (a) CBCT imaging device, captured at the Department of Stomatology, First Affiliated Hospital of Nankai University; (b) The fundamental principle of CBCT scanning involves a rotating gantry, an X-ray tube, and an image detector kaasalainen2021dental.
  • Figure 5: The construction steps of a digital dental model: (a) Intraoral scanner; (b) The intraoral scanner captures raw data, stored in point cloud format (common file extensions such as .ply or .xyz); (c) Mesh model (typically in formats like .obj or .stl); (d) Digital dental model.
  • ...and 19 more figures