Optimum display of the digital intraoral imaging system by quantitative evaluation on image quality

2005 Fiscal Year Final Research Report Summary

• Project/Area Number: 16591888
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Pathobiological dentistry/Dental radiology
• Research Institution: KYUSHU UNIVERSITY
• Principal Investigator: YOSHIURA Kazunori Kyushu University, Faculty of Dental Science, Professor, 歯学研究院, 教授 (20210643)
• Co-Investigator(Kenkyū-buntansha): KAWAZU Toshiyuki Kyushu University, Kyushu University Hospital, Research Associate, 大学病院, 助手 (20294960) ; CHIKUI Toru Kyushu University, Kyushu University Hospital, Assistant Professor, 大学病院, 講師 (10295090) ; SHIMIZU Mayumi Kyushu University, Faculty of Dental Science, Research Associate, 歯学研究院, 助手 (50253464) ; TOKUMORI Kenji Kyushu University, Faculty of Dental Science, Research Associate, 歯学研究院, 助手 (40253463) ; YOSHINAKA Masanori Kyushu University, Kyushu University Hospital, Radiological Technologist, 大学病院, 診療放射線技師 (80380509)
• Project Period (FY): 2004 – 2005
• Keywords: digital intraoral imaging system / optimum display / phantom / low contrast detectability / diagnostic accuracy / image contrast

Research Abstract

Digital system is widely used in medical imaging and this trend is found in general dental practitioners, too. However, no consensus was obtained about the optimum display of the digital image and information in the digital images are not fully used for diagnosis and treatment planning.
In this study, low contrast detectability in the digital systems was compared with that in the film using specially developed image quality phantom for quantitative evaluation. Then diagnostic accuracy for proximal caries of the extracted teeth was evaluated using the same systems and the same observers.
Regression analysis showed high correlation between low contrast detectability and diagnostic accuracy. There was no statistically significant difference between diagnostic accuracy of the digital systems and that of the film. Effect of exposure on diagnostic accuracy or low contrast detectability, however, was larger in the digital systems than in the film. This effect was smaller in the digital system with automatic exposure compensation than in that without it.
Further investigation clarified that image contrast of the digital system was completely different from that of the film. Image contrast of the digital system in low exposure was far inferior to that of the film. We developed the image display algorithm to improve contrast in low exposure. The result showed that overall image contrast of the processed image was better than that of the original and it was similar to that of the film.
This image display algorithm may have general applicability to all digital intraoral imaging systems.