RSNA 2013 

Abstract Archives of the RSNA, 2013


LL-PHS-MO6B

Optimal Reconstruction Filter on Observation of Periodontal Ligament Space in Dental Cone-beam CT

Scientific Informal (Poster) Presentations

Presented on December 2, 2013
Presented as part of LL-PHS-MOB: Physics - Monday Posters and Exhibits (12:45pm - 1:15pm)

Participants

Yuuki Houno, Presenter: Nothing to Disclose
Toshimitsu Hishikawa, Abstract Co-Author: Nothing to Disclose
Ken-Ichi Gotoh, Abstract Co-Author: Nothing to Disclose
Munetaka Naitoh, Abstract Co-Author: Nothing to Disclose
Eiichiro Ariji, Abstract Co-Author: Nothing to Disclose
Yoshie Kodera PhD, Abstract Co-Author: Nothing to Disclose

PURPOSE

Diagnosis of the alveolar bone condition is important in the treatment of periodontal disease. However, visualization of the thin alveolar bone and periodontal ligament space is difficult owing to image blurring using the current imaging technique. Therefore, identification of the marginal alveolar bone level is unreliable. In this study, we try to determine high quality images of periodontal ligament space separation were obtained by changing the reconstruction filter.

METHOD AND MATERIALS

Images from 4 different areas of 2 types of head phantoms were obtained using Alphard-3030 dental cone-beam computed tomography (Asahi Roentgen Industry Co., Ltd.). The image was reconstructed from the raw data by using 5 different reconstruction filters. Post reconstruction, these reconstructed images were compared with the Alphard-3030 reconstructed images by using Thurstone’s paired comparison, and a scale value was calculated. The images were evaluated limiting the clarity of periodontal ligament space boundary by 3 periodontists and 3 radiologists with >5 years of clinical experience. To confirm the image characteristics, modulation transfer function and Wiener spectrum were calculated.

RESULTS

The best image was one reconstructed with filter that increased the high-frequency component based on Shepp and Logan’s function. The Alphard-3030 reconstructed image, which is used in clinical settings, was the worst. The image reconstructed with the best filter had a higher modulation transfer factor and Wiener spectral value than the Alphard-3030 reconstructed image. These results suggest that sharp images enhance the visualization of the periodontal ligament space, and the imaging noise is insignificant. These images enable the observer to clearly view the periodontal ligament space boundary.

CONCLUSION

We determined the optimal reconstruction filter required to observe the periodontal ligament space. 

CLINICAL RELEVANCE/APPLICATION

Image quality of periodontal ligament space may be improved by alternating reconstruction filter of cone-beam CT. This method can be applied to various CT apparatus without hardware replacement.

Cite This Abstract

Houno, Y, Hishikawa, T, Gotoh, K, Naitoh, M, Ariji, E, Kodera, Y, Optimal Reconstruction Filter on Observation of Periodontal Ligament Space in Dental Cone-beam CT.  Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL. http://archive.rsna.org/2013/13044170.html