Dagfinn Svanaes

Intraoral storage phosphor radiography for approximal caries detection and effect of image magnification : Comparison with conventional radiography

OBJECTIVES The aim of this study was first to test the newly developed storage phosphor (SP) system Digora for diagnostic accuracy of approximal caries detection with special emphasis on image magnification compared with conventional radiography, and second, to test whether the image surroundings could have an influence on observer performance. DESIGN SP plates and Ektaspeed films were exposed simultaneously to 50 extracted premolars/molars, and the SP image files were transported to a different platform for processing. Ten observers rated 95 approximal surfaces for caries with four imaging modalities: (1) SP images at dental film size viewed with black surroundings, (2) magnified (x4) images viewed in the same manner, (3) magnified (x4) images viewed with text and graphics framing, and (4) films viewed on a light box. The observations were validated with microscopy. Diagnostic accuracy was calculated as the area beneath the receiver operating characteristic curve (AZ). RESULTS No significant differences were found between SP images and films. Magnified SP images with text and graphics demonstrated significantly higher AZ values than nonmagnified images both for enamel and dentin, and magnified images with black surroundings demonstrated similar results for dentin. CONCLUSIONS In this in vitro material the observer performance for approximal caries detection with SP images was comparable to that with Ektaspeed films. The magnified SP images generally showed a significantly higher diagnostic accuracy than the nonmagnified SP images.

Reproducibility of radiographs with the orthopantomograph 5: tooth-length assessment.

The precision of tooth-length assessments based on repeated panoramic radiographs made with a Siemens OP 5 was investigated in three groups of twenty patients. Two exposures of the same patient were made under three different radiographic conditions. Of the tooth lengths, nonmeasurability was found in 14% to 17%. The variability (standard deviation) of the measurements assessed from repeated radiographs ranged from 0.65 to 0.85 mm, or 2.4% to 3.1% of the mean radiographic tooth length in the different patient groups. The measurement error ranged from 0.43 to 0.56 mm, indicating that the main source of error inherent in the method was recognition of the reference points. Small differences were found between the tooth groups and between the right and left sides.

Approximal Caries Depth Assessment with Storage Phosphor versus Film Radiography

Objectives: To study whether caries–specific enhancement of storage phosphor images might improve the observer performance of approximal caries depth assessments compared with film radiography. Materials and Methods: 120 exposures were made of 120 extracted human teeth. To obtain geometrically identical images, Ektaspeed Plus films and storage phosphor plates were exposed simultaneously. The imaging plates were scanned in a Digora® scanner and the files transferred to a different platform for image enhancement. Nine observers viewed films and storage phosphor images without provisions for adjustment of image intensity and contrast. For each imaging modality, 240 approximal surfaces were rated for caries on a 5–point confidence scale. Definite and probable caries lesions were also rated for lesion depth, and all ratings were compared with the histological state. Diagnostic accuracy was expressed as the area under the ROC curve (Az value). Paired t tests were used to compare the imaging modalities for diagnostic accuracy and F tests to compare observer variances. Results: Enhanced storage phosphor images demonstrated significantly higher mean Az values than film (p = 0.0066). Significantly higher mean Az values were demonstrated in the outer half of enamel (p = 0.01), but no significant differences were found between the modalities for caries lesions penetrating beyond the outer half of the enamel. The number of correctly diagnosed true–positive surfaces with caries in outer enamel was significantly higher with storage phosphor images than with film (p = 0.00014). False–positive surfaces were most frequently registered in the outer enamel with both modalities, but in this region the number of false–positive surfaces was significantly higher with storage phosphor images than with film (p = 0.0038). Pooled sensitivity and specificity values were 0.48/0.94 and 0.61/0.86 for film and storage phosphor images, respectively. The interobserver variability was significantly lower for storage phosphor images than for film. Conclusion: Enhancement of storage phosphor images with a caries–specific procedure significantly improved the accuracy of caries depth assessments in the outer half of the enamel compared with film radiography and reduced observer variability.