RSNA 2011 

Abstract Archives of the RSNA, 2011


LL-INS-WE2A

A Fully Automatic Image Analysis System for Pathologic Signs on Panoramic Radiographs: Integration of Computer-aided Detection (CAD) Schemes into Dental PACS System

Scientific Informal (Poster) Presentations

Presented on November 30, 2011
Presented as part of LL-INS-WE: Informatics

Participants

Akitoshi Katsumata DDS, PhD, Presenter: Nothing to Disclose
Tatsuro Hayashi PhD, Abstract Co-Author: Nothing to Disclose
Tsuyoshi Sawagashira BS, Abstract Co-Author: Nothing to Disclose
Takuya Matsumoto BS, Abstract Co-Author: Nothing to Disclose
Takeshi Hara PhD, Abstract Co-Author: Nothing to Disclose
Hiroshi Fujita PhD, Abstract Co-Author: Research grant, Hitachi, Ltd
Motoki Tagami, Abstract Co-Author: Nothing to Disclose
Chisako Muramatsu PhD, Abstract Co-Author: Nothing to Disclose
Xiangrong Zhou PhD, Abstract Co-Author: Nothing to Disclose
Yukihiro Iida, Abstract Co-Author: Nothing to Disclose
Masato Matsuoka, Abstract Co-Author: Nothing to Disclose
Kiyoji Katagi DDS, PhD, Abstract Co-Author: Nothing to Disclose

CONCLUSION

In order to build a new screening pathway that identifies asymptomatic patients, the PACS-CAD system for detecting pathologic signs on panoramic radiographs was developed. The experimental results suggested that our system had the potential to identify asymptomatic patients.

BACKGROUND

Findings of dental panoramic radiographs are including pathologic signs that are suggestive of possible osteoporosis, strokes, and maxillary sinusitis. To design a new screening pathway via dentists has a potential to accelerate early detection of pathologic conditions. However, general dentists are unfamiliar with such pathologic signs on panoramic images. Furthermore, to add supplemental readings may be seen as a time-consuming interruption by the dentists in their busy schedule.

EVALUATION

We have implemented three automatic CAD schemes on panoramic images; (1) measurement of the mandibular cortical thickness for osteoporosis, (2) detection of calcifications in the cervical region for strokes, (3) evaluation of the radiopacity in bilateral maxillary sinuses for maxillary sinusitis. Then, they were integrated into the dental PACS system. In our system, when dentists input panoramic images into the dental PACS, it sends images to the CAD schemes and receives CAD results automatically. Therefore, the CAD results are available without user interactions. To evaluate the clinical usefulness of our system, a preliminary clinical trial at one dental clinic is currently in progress. Until now, some asymptomatic patients with osteoporosis were detected by our system.

DISCUSSION

To detect pathologies in early stages is very important to improve patient's prognosis. However, the asymptomatic patients do not always visit the medical clinic to do the medical checkup. Panoramic radiographs including various pathologic signs are widely used in clinical dentistry. Therefore, a supplemental screening pathway in dentistry using panoramic images was expected as a tool for identifying asymptomatic patients. To achieve such a potential need of panoramic images, the PACS-CAD system for detecting pathologic signs was developed. The clinical utility of our system will be demonstrated by the result of the preliminary clinical trial.

Cite This Abstract

Katsumata, A, Hayashi, T, Sawagashira, T, Matsumoto, T, Hara, T, Fujita, H, Tagami, M, Muramatsu, C, Zhou, X, Iida, Y, Matsuoka, M, Katagi, K, A Fully Automatic Image Analysis System for Pathologic Signs on Panoramic Radiographs: Integration of Computer-aided Detection (CAD) Schemes into Dental PACS System.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11011028.html