Grand Challenges in Dental X-ray Image Analysis

Ching-Wei Wang
National Taiwan University of Science and Technology, Taiwan

Dr Cheng-Ta Huang
National Taiwan University of Science and Technology, Taiwan

Chung-Hsing Li
Tri-Service General Hospital, Taiwan

Sheng-Wei Chang
Tri-Service General Hospital, Taiwan

Website
Link

Challenge Abstract

Two independent sub-challenges are proposed, including automated detection and statistical analysis for diagnosis in cephalometric X-ray image and computer-automated detection of caries in bitewing radiography.

Challenge #1: Automated Detection and Analysis for Diagnosis in Cephalometric X-ray Image
Cephalometric analysis is mandatory, providing the interpretation of patients’ bony, dental and soft tissue structures and giving the whole pictures for the orthodontic analysis and treatment plan making. In clinical practice, manual marking is commonly conducted, and the analysis of cephalogram is frequently performed by dentists, orthodontists, and oral and maxillofacial surgeons in treatment planning. However, the procedure of manual marking and analysis of cephalogram is very time consuming and subjective. Automated methods for anatomical landmark detection and identification of anatomical abnormalities could be the solution to facilitate these issues. However, automated anatomical landmark detection of cephalometric x-ray images is challenging. From literature review, in recent 7 years the success rate of automatic landmarks has increased from 71% to only 89.5% with the 2.0 mm precision range, which is the accepted precision range of a landmark in clinical practice. In 2014, we held an automatic cephalometric X-ray landmark detection challenge in ISBI 2014, Beijing, China. 300 cephalometric X-ray images were used in ISBI-2014 challenge, and the best detection result for 19 landmarks can be achieved up to 71.48% using the 2 mm precision range. In this challenge, a larger clinical database is built, and cephalometric x-ray images of 500 patients with ground truth data produced by two experienced medical doctors are collected. In addition, apart from the landmark detection task, a further task in identifying anatomical abnormalities to assist clinical diagnosis using these landmarks is added in this new challenge.

The challenge participating teams will be invited to contribute to a joint journal paper describing and summarizing the challenge outcome, which will be submitted to a high-impact SCI journal in the field.

Challenge #2: Computer-Automated Detection of Caries in Bitewing Radiography Dental caries is a transmissible bacterial disease of the teeth that would destructs the structure of teeth. The dentist has approached diagnosing and treating dental caries based mostly on radiographs. While dental caries is a disease process, the term is routinely used to describe radiographic radiolucencies. Radiographic examination can improve detection and diagnostic of the earliest sign of dental caries’ demineralization. Automated caries lesion detection technologies provide potential diagnostic data for dental practitioners and assist identifying signs of various diseases. However, accurate and objective methods for radiographic caries diagnosis are poorly explored. Therefore, the aim of this challenge is to investigate possible automated methods for detection of caries in bitewing radiography.

The challenge participating teams will be invited to contribute to a joint journal paper describing and summarizing the challenge outcome, which will be submitted to a high-impact SCI journal in the field.

About wpengine

This is the "wpengine" admin user that our staff uses to gain access to your admin area to provide support and troubleshooting. It can only be accessed by a button in our secure log that auto generates a password and dumps that password after the staff member has logged in. We have taken extreme measures to ensure that our own user is not going to be misused to harm any of our clients sites.