Computerized numerical evaluation of radiographic images: the destruction and reduction of bone tissues in periodontal areas

Image processing for enhanced observer agreement in the evaluation of periapical bone changes

AIM

The aim of the present study was to evaluate the effect of root canal treatment on periapical lesions by conventional and subtracted digital radiographic images of clinical cases.

METHODOLOGY

Eleven patients who exhibited clinical or radiological signs of periapical pathology received root canal treatment. Periapical radiographs were obtained immediately postoperatively and recall radiographs at intervals of 3, 6, 9 and 12 months postoperatively were obtained. Identical exposure geometry was maintained. From the standardized radiographs digitized images were produced. Four experienced practitioners interpreted the radiographs and the digitized images. They assessed 59 pairs of images projected at random using a slide projector. On the left side, there was a reference image with no lesion, and on the right an image either with or without a lesion. Each reader was asked to rate each pair of images on a three-point scale: yes, absolutely sure that gain or loss was present; uncertain, if there was gain or loss; no, absolutely sure that no gain or loss was present. There was no time limit for each decision. Inter- and intra-examiner agreement was analyzed using the kappa-statistic for the diagnosis of periapical bone density changes in the periapical region at the different time points either in conventional pairs of radiographs or using digital subtraction images.

RESULTS

The inter-examiner agreement (P < 0.001) and the intra-examiner agreement (P = 0.02) when digital pictures where evaluated were significantly higher than with conventional radiographs.

CONCLUSIONS

A highly significantly better observer agreement was achieved by digital subtraction radiography during the evaluation of the outcome of root canal treatment on periapical lesions.

Color conversion of alveolar bone density changes in digital subtraction images

This study evaluated the influence of image processing of digital subtraction images on inter- and intra-examiner agreement relative to the detection of alveolar bone change on radiographs. 62 sites on digital subtraction images from standardized radiographs from patients requiring periodontal surgery were included in this study. The images were displayed as digital subtraction images, contrast enhanced subtraction images and as color-converted digital subtraction images. The images were projected in random order, analyzed by 4 interpreters and 1 of 8 possible diagnostic decisions was noted for each one of 62 sites. 2 months after the first evaluation, the slides were re-analyzed. 3 different computations (overall agreement, specific agreement and kappa-value) were calculated. The results indicated that image processing of subtraction images using pseudocolor display of the density changes might improve the intra- and inter-examiner agreement.

Digital imaging in periodontal radiography. A review

The development and the potential impact of digital imaging techniques for the analysis of dental radiographs are reviewed with emphasis on periodontal diagnosis. The introduction of video-based computer-assisted image analysis systems for the evaluation of standardized dental radiographs offered increased diagnostic accuracy in the detection of subtle density changes over interpretation of conventional radiographs. Digital subtraction images were interpreted qualitatively on a video monitor or quantitatively by applying programs which measure the density differences stored in digital format. Recently, these methods, which have been extensively tested in vitro, were applied to clinical trials for the assessment of alveolar bone density changes due to progression of disease or healing following therapy. The application of digital image analysis methods offers objective, quantitative and non-invasive ways to obtain additional diagnostic information from standardized radiographs in clinical trials.

Computer-assisted densitometric image analysis in periodontal radiography. A methodological study

A videobased computer assisted densitometric image analysis (CADIA) system to quantify alveolar bone density changes on standardized dental radiographs was tested. An algorithm was used for grey level correction of a subsequent image to the baseline image. Quantitative information regarding positive and/or negative grey level changes were obtained automatically. Comparison of the ability of CADIA to detect surgically induced bone loss with interpretation of digital subtraction images and conventional radiographic interpretation revealed that CADIA was the most sensitive of the 3 methods, followed by interpretation of digital subtraction images which was considerably more sensitive than conventional radiographic interpretation. CADIA was capable of assessing differences in alveolar bone changes due to periodontal surgery between sites exposed to ostectomy/osteoplasty and control sites and sites exposed to periodontal surgery without ostectomy/osteoplasty. Finally, CADIA was capable of assessing differences in remodeling activity over 4-6 weeks after periodontal surgery between 45 surgical sites and 45 control sites. The system offers an objective method to quantitatively follow alveolar bone density changes over time and appears to be the most sensitive of previously described radiographic interpretation techniques.

Detection and measurement of approximal radiolucencies by computer-aided image analysis

Serial standardized bitewing radiography permits monitoring of the development of approximal carious lesions. A quantitative method using image-analysis techniques was developed. Only certain types of hardware proved capable of processing this complex image adequately. A computer-aided, software-driven, TV-based system incorporating an image memory, recursive filtration, and 256 gray-level resolution, which can both detect and measure approximal enamel radiolucencies produced by natural carious lesions, has been developed. The radiolucent boundary is delineated, displayed, and plotted to scale. For the first time, objective values for depth, area, width, perimeter, and "degree of radiolucency" can be automatically derived. Currently the system is seen as a tool in caries research; in the future such systems may have wider applications.