Visual linearization of image data for the display of digital intraoral radiographs

Digital intraoral radiography has developed rapidly during the past decade. The most important advantage of this technique is that it breaks the fixed relationship between exposure and density that is inherent in film radiography. Thus, processing of digital intraoral radiographs after image capturing has become possible. A great number of digital image processing algorithms have been introduced. However, these algorithms do not directly consider the exponential attenuation of an exposed object, which has an essential impact on image quality. Furthermore, the response of the human visual system has been considered only in a limited number of contributions.

New digital image processing algorithms have been derived that take into account both the exponential attenuation when a radiograph is exposed and the response of the human visual system. By applying these algorithms, equal steps in object thickness will be perceived as equal steps in brightness when a digital intraoral radiograph is viewed on a computer monitor.

To determine if these algorithms improve perception of small contrasts and diagnostic accuracy, a series of studies was performed by means of the so-called Perceptibility Curve (PC) test, by means of in vitro models and by means of an in vivo test employing clinical digital intraoral radiographs.

A PC test showed that the perception of small contrasts was improved in both 8 bit and 12 bit digital intraoral radiographs processed with the new algorithms.

An in vitro study demonstrated that the diagnostic accuracy of approximal carious lesions was improved in digital intraoral radiographs processed with the new algorithms.

An additional in vitro study demonstrated that radiographs processed with the new algorithms may improve the determination of the length of thin endodontic files. Such processed radiographs were essentially comparable with radiographs processed with the default processing method employed by one commercially available digital intraoral radiographic system. One study was performed employing clinical intraoral radiographs. In this study, an evaluation of the subjective diagnostic image quality was performed employing radiographs processed with three different algorithms. The results indicated that radiographs processed with the new algorithms were favorable.

In conclusion, application of the new algorithms presented in this work may improve perception of small contrasts and may also improve diagnostic accuracy.

List of scientific papers

I. Li G, Yoshiura K, Welander U, Sallstrom P, McDavid WD (2002). Visual linearization of the display of digital radiographs. Dentomaxillofac Radiol. 31(2): 131-6.
https://pubmed.ncbi.nlm.nih.gov/12076054

II. Welande U, Yoshiura K, Li G, Sallstrom P, McDavid WD (2002). Correction for attenuation and visual response in digital radiography. Dentomaxillofac Radiol. 31(2): 117-25.
https://pubmed.ncbi.nlm.nih.gov/12076052

III. Li G, Welander U, Yoshiura K, Shi XQ, McDavid WD (2003). Perceptibility curve test for digital radiographs before and after correction for attenuation and correction for attenuation and visual response. [Accepted]

IV. Li G, Yoshiura K, Welander U, Shi XQ, McDavid WD (2002). Detection of approximal caries in digital radiographs before and after correction for attenuation and visual response. An in vitro study. Dentomaxillofac Radiol. 31(2): 113-6.
https://pubmed.ncbi.nlm.nih.gov/12076051

V. Li G, Sanderink GCH, Welander U, McDavid WD, Nasstrom K (2004). Evaluation of endodontic files in digital radiographs before and after employing three image processing algorithms. Dentomaxillofac Radiol. 33.

VI. Li G (2004). Comparative investigation of subjective image quality of digital intraoral radiographs processed with three image-processing algorithms. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 97. [Accepted]