Comparison of standard and task-specific enhancement of Digora storage phosphor images for approximal caries diagnosis

Reduction of scratch or dirt artifacts on intraoral radiographs using dual imaging plates in image processing

OBJECTIVES

Artifacts including scratches and dirt artifacts on the digital intraoral radiographs finally contribute to making inaccurate diagnoses. The aim of this study was to reduce the incidence of artifacts using dual imaging plates (DIPs) in imaging processing.

METHODS

Conventional X-rays were taken of a porcine mandible embedded in acrylic resin using a DIP which consists of a front IP (FIP) and a back IP (BIP) with some scratches and dirt. The two images of the FIP and BIP were then synthesized and averaged to obtain a conventional DIP image. The following image processing method was used to make a DIP with artifact reduction (DIP+AR) image. A subtraction image of the FIP and BIP was constructed and the standard deviation (SD) was calculated. If the pixel value was over 3SD on the subtraction images, the pixel value of the DIP was swapped with the value on the opposite side of the non-artifact pixel. The conventional and DIP+AR images were also subjectively evaluated.

RESULTS

Image processing to create a DIP+AR image was able to reduce the number of artifacts. Medians of number of artifacts evaluated were 2.00 [interquartile range (IQR), 2.50] in DIP images and 0.67 (IQR, 1.29) in DIP+AR images, indicating a significant reduction of number of artifacts in DIP+AR images.

CONCLUSIONS

DIP+AR image processing can reduce the incidence of artifacts caused by scratches and dirt, and could extend the lifespan of the IP and contribute accurate diagnosis in oral radiology.

Tooth-Related Disease Detection System Based on Panoramic Images and Optimization Through Automation: Development Study

Background

Early detection of tooth-related diseases in patients plays a key role in maintaining their dental health and preventing future complications. Since dentists are not overly attentive to tooth-related diseases that may be difficult to judge visually, many patients miss timely treatment. The 5 representative tooth-related diseases, that is, coronal caries or defect, proximal caries, cervical caries or abrasion, periapical radiolucency, and residual root can be detected on panoramic images. In this study, a web service was constructed for the detection of these diseases on panoramic images in real time, which helped shorten the treatment planning time and reduce the probability of misdiagnosis.

Objective

This study designed a model to assess tooth-related diseases in panoramic images by using artificial intelligence in real time. This model can perform an auxiliary role in the diagnosis of tooth-related diseases by dentists and reduce the treatment planning time spent through telemedicine.

Methods

For learning the 5 tooth-related diseases, 10,000 panoramic images were modeled: 4206 coronal caries or defects, 4478 proximal caries, 6920 cervical caries or abrasion, 8290 periapical radiolucencies, and 1446 residual roots. To learn the model, the fast region-based convolutional network (Fast R-CNN), residual neural network (ResNet), and inception models were used. Learning about the 5 tooth-related diseases completely did not provide accurate information on the diseases because of indistinct features present in the panoramic pictures. Therefore, 1 detection model was applied to each tooth-related disease, and the models for each of the diseases were integrated to increase accuracy.

Results

The Fast R-CNN model showed the highest accuracy, with an accuracy of over 90%, in diagnosing the 5 tooth-related diseases. Thus, Fast R-CNN was selected as the final judgment model as it facilitated the real-time diagnosis of dental diseases that are difficult to judge visually from radiographs and images, thereby assisting the dentists in their treatment plans.

Conclusions

The Fast R-CNN model showed the highest accuracy in the real-time diagnosis of dental diseases and can therefore play an auxiliary role in shortening the treatment planning time after the dentists diagnose the tooth-related disease. In addition, by updating the captured panoramic images of patients on the web service developed in this study, we are looking forward to increasing the accuracy of diagnosing these 5 tooth-related diseases. The dental diagnosis system in this study takes 2 minutes for diagnosing 5 diseases in 1 panoramic image. Therefore, this system plays an effective role in setting a dental treatment schedule.

Radiographic modalities for diagnosis of caries in a historical perspective: from film to machine-intelligence supported systems

Radiographic imaging for the diagnosis of caries lesions has been a supplement to clinical examination for approximately a century. Various methods, and particularly X-ray receptors, have been developed over the years, and computer systems have focused on aiding the dentist in the detection of lesions and in estimating lesion depth. The present historical review has sampled accuracy ex vivo studies and clinical studies on radiographic caries diagnosis that have compared two or more receptors for capturing the image. The epochs of film radiography, xeroradiography, digital intraoral radiography, panoramic radiography and other extraoral methods, TACT analysis, cone-beam CT and artificial intelligence systems aiding in decision-making are reviewed. The author of this review (43 years in academia) has been involved in caries research and contributed to the literature in all the mentioned epochs.

Evaluation of a Mathematical Model for Digital Image Enhancement

Objective : The purpose of this study is to compare the detected number of holes on a stepwedge on images resulting from the application of the 5th degree polynomial model compared to the images resulting from the application of linear enhancement. Material and Methods : A 10-step aluminum step wedge with holes randomly drilled on each step was exposed with three different kVp and five exposure times per kVp on a Schick33^® sensor. The images were enhanced by brightness/contrast adjustment, histogram equalization and with the 5th degree polynomial model and compared to the original non-enhanced images by six observers in two separate readings. Results : There was no significant difference between the readers and between the first and second reading. There was a significant three-factor interaction among Method, Exposure time, and kVp in detecting holes. The overall pattern was: “Poly” results in the highest counts, “Original” in the lowest counts, with “B/C” and “Equalized” intermediate. Conclusion : The 5th degree polynomial model showed more holes when compared to the other modalities.

The efficacy of the reverse contrast mode in digital radiography for the detection of proximal dentinal caries

Purpose

This study evaluated the diagnostic accuracy of the reverse contrast mode in intraoral digital radiography for the detection of proximal dentinal caries, in comparison with the original digital radiographs.

Materials and Methods

Eighty extracted premolars with no clinically apparent caries were selected, and digital radiographs of them were taken separately in standard conditions. Four observers examined the original radiographs and the same radiographs in the reverse contrast mode with the goal of identifying proximal dentinal caries. Microscopic sections 5 µm in thickness were prepared from the teeth in the mesiodistal direction. Four slides prepared from each sample used as the diagnostic gold standard. The data were analyzed using SPSS (α=0.05).

Results

Our results showed that the original radiographs in order to identify proximal dentinal caries had the following values for sensitivity, specificity, positive predictive value, negative predictive value, and accuracy, respectively: 72.5%, 90%, 87.2%, 76.5%, and 80.9%. For the reverse contrast mode, however, the corresponding values were 63.1%, 89.4%, 87.1%, 73.5%, and 78.8%, respectively. The sensitivity of original digital radiograph for detecting proximal dentinal caries was significantly higher than that of reverse contrast mode (p<0.05). However, no statistically significant differences were found regarding specificity, positive predictive value, negative predictive value, or accuracy (p>0.05).

Conclusion

The sensitivity of the original digital radiograph for detecting proximal dentinal caries was significantly higher than that of the reversed contrast images. However, no statistically significant differences were found between these techniques regarding specificity, positive predictive value, negative predictive value, or accuracy.

Detection of artificial demineralization bordering different types of laminate veneers using visual inspection and storage phosphor radiography

OBJECTIVE

The objective of this study is to compare the diagnostic accuracy of visual inspection (VI) and storage phosphor plate (SPP) radiography for the detection of artificial demineralization bordering different laminate veneers.

MATERIALS AND METHODS

Twenty human maxillary canine teeth were prepared. All-ceramic (A) and hybrid ceramic (H) laminate veneers were fabricated and luted. Veneered teeth were covered except for a circular window on the proximal surface bordering restorations. Teeth were kept in acetic acid buffer to create demineralization and imaged with a SPP system. Ten observers evaluated all teeth first visually then with SPP images for the presence/absence of demineralization. Teeth were examined using scanning electron microscopy (SEM) as well. The accuracy was expressed as the area under the ROC curves (A(z)). Pair-wise comparisons were performed using two-way ANOVA and post hoc t test (p = 0.05). Fleiss kappa (κ) was used for agreement.

RESULTS

SPP radiography was better than the VI for both veneers (p = 0.004). The A(z)s of two veneers were different for both VI (p < 0.005) and SPP (p < 0.005). SEM evaluation revealed lesions confined to enamel. κ was fair for H, and fair to moderate for A. Agreement was higher for the radiographic evaluation for both veneers.

CONCLUSION

Enamel demineralizations bordering hybrid and ceramic laminate veneers can be detected better with SPP radiography than VI and detectability was better for all-ceramic veneers than the hybrid ceramic ones.

CLINICAL RELEVANCE

Early detection of enamel demineralizations bordering laminate veneers would result in time-saving and less-invasive treatment methods; therefore, SPP radiography may be recommended in clinically suspicious cases since it provides better diagnostic accuracy.

Detection of vertical root fractures using digitally enhanced images: reverse-contrast and colorization

Root fracture diagnosis is a clinical difficulty that in most cases can only be detected through radiography. The objective of this study was to compare the diagnostic accuracy of two types of digitally enhanced images (reverse-contrast and colorization) with original digital radiographies in detecting experimental root fractures. Two hundred extracted single-rooted human teeth were endodontically instrumented and then divided in two groups, one control group and one test group, including fractured teeth. Vertical root fractures were experimentally made in the fractured group. The digital image of each tooth was taken, using the paralleling technique. There were three groups of images: (i) original, (ii) reverse contrast, and (iii) colorized. Three experienced dental specialists examined the images with no prior knowledge of the distribution of the root fractures. Two-way analysis of variance was used to assess the differences in accuracy, sensitivity, and specificity of each technique in detecting root fractures. Cohen's kappa coefficients were calculated to investigate the degree of interobserver agreement. The accuracy, sensitivity, and specificity of original images were 67.4%, 66.7%, and 68%, respectively; these amounts were 61.5%, 61%, and 65.5% in reverse contrast images and 66.4%, 70.7%, and 62% in colorized digital radiography. The original images had the best inter observer kappa coefficients (between 0.45 and 0.55). The results showed that the accuracy of original images is better than reverse contrast and colorized images. Use of reverse-contrast and colorization digital images in root fracture detection should be regarded as an adjunct to other diagnostic methods not as a highly critical diagnostic aid.

Diagnostic accuracy of proximal enamel subsurface demineralization and its relationship with calcium loss and lesion depth

OBJECTIVE

The aim of this study was to determine the relationship between the amount of calcium loss, lesion depth, and the accuracy of storage phosphor plate (SPP) and film radiographs for the detection of artificial proximal demineralization.

METHODS

Standard enamel windows of extracted premolars were exposed to a demineralizing solution for 60 h, 80 h, 100 h and 120 h. Solutions were analysed for calcium concentration by atomic absorption spectrometer and the lesion depths were calculated by a specific formula. All teeth were radiographed with SPPs and F-speed films before and after acid application. Images were evaluated by five observers. Stereomicroscopic and scanning electron microscopic (SEM) observations were carried out to visualize enamel surfaces after acid exposure. Receiver operating characteristic analysis was used for diagnostic accuracy (A(z)). A(z)s were compared with factorial analysis of variance and t-tests. The relationship between A(z)s and lesion depths was determined with Pearson's correlation test.

RESULTS

Strong positive correlation was found between A(z)s of both radiographic methods and lesion depths. No difference was found between the A(z)s of two radiographic systems for any of the demineralization durations (p > 0.05). Pair-wise comparisons revealed no significant difference in A(z)s of SPPs (p > 0.05), while significant differences were obtained for the A(z)s of films for different demineralization periods (p < 0.05). Stereomicroscopic and SEM observations confirmed demineralizations from superficial to deeper layers of enamel.

CONCLUSION

Subsurface enamel demineralization was not accurately detectable with either storage phosphor plates or F-speed films. The amount of calcium loss and the depth of demineralization have a strong relationship with diagnostic accuracy with a significant effect particularly on F-speed films.

Accuracy of direct digital radiography for detecting occlusal caries in primary teeth compared with conventional radiography and visual inspection: an in vitro study

OBJECTIVES

the diagnosis of caries lesions is still a matter of concern in dentistry. The diagnosis of dental caries by digital radiography has a number of advantages over conventional radiography; however, this method has not been explored fully in the field of paediatric dentistry. This in vitro research evaluated the accuracy of direct digital radiography compared with visual inspection and conventional radiography in the diagnosis of occlusal caries lesions in primary molars.

METHODS

50 molars were selected and evaluated under standardized conditions by 2 previously calibrated examiners according to 3 diagnostic methods (visual inspection, conventional radiography and direct digital radiography). Direct digital radiographs were obtained with the Dixi3 system (Planmeca, Helsinki, Finland) and the conventional radiographs with InSight film (Kodak Eastman Co., Rochester, NY). The images were scored and a reference standard was obtained histologically. The interexaminer reliability was calculated using Cohen's kappa test and the specificity, sensitivity and accuracy of the methods were calculated.

RESULTS

examiner reliability was good. For lesions limited to the enamel, visual inspection showed significantly higher sensitivity and accuracy than both radiographic methods, but no significant difference was found in specificity. For teeth with dentinal caries, no significant differences were found for any parameter when comparing visual and radiographic evaluation.

CONCLUSIONS

although less accurate than the visual method for detecting caries lesions confined to the enamel, the direct digital radiographic method is as effective as conventional radiographic examination and visual inspection of primary teeth with occlusal caries when the dentine is involved.

Modern dental imaging: a review of the current technology and clinical applications in dental practice

A review of modern imaging techniques commonly used in dental practice and their clinical applications is presented. The current dental examinations consist of intraoral imaging with digital indirect and direct receptors, while extraoral imaging is divided into traditional tomographic/panoramic imaging and the more recently introduced cone beam computed tomography. Applications, limitations and current trends of these dental "in-office" radiographic techniques are discussed.

The effects of digital image enhancement on the detection of vertical root fracture

AIM

To determine the effects of digital image enhancement on observer ability to detect experimentally induced vertical root fractures (VRF).

MATERIAL AND METHODS

A total of 64 extracted human mandibular premolar teeth were used in this study. In 32 teeth, VRFs were created in the bucco-lingual planes by gently tapping with screw-type root-canal pins. The remaining 32 intact teeth served as a control group. Digital images were obtained using a charge coupled device sensor. Three observers separately examined the original and four types of digitally enhanced images (enhanced using sharpness, zoom-in, reverse-contrast, and pseudo-3D functions) at 1-week intervals. All teeth were evaluated using a 5-point scale for the presence/absence of VRF. Evaluations of each image set were repeated 1 month after the initial viewings. Kappa coefficients were calculated to investigate the degree of intra- and inter-observer agreement. The areas under the receiver operating characteristic (ROC) curves (Az values) were calculated using the MedCalc statistical software. ROC values for each image type, observer and viewing were compared using t-tests. A level of alpha = 0.05 was considered significant.

RESULTS

Kappa coefficients for intra-observer agreement ranged from 0.304 to 0.679. Inter-observer agreement kappa values ranged from 0.109 to 0.399 for the first reading and from 0.106 to 0.380 for the second reading. Statistical comparisons between Az values for each observer showed no significant differences (P > 0.05) among image types.

CONCLUSION

There were no differences in diagnostic outcomes among differently enhanced images in the in vitro detection of VRF.

Task-specific enhancement filters in storage phosphor images from the Vistascan system for detection of proximal caries lesions of known size

OBJECTIVE

The aim was to compare original 8-bit images from the Vistascan storage phosphor plate system with images enhanced with specific filters for detection of proximal caries lesions and, further, to analyze the mutual sensitivity of the filtered images for different lesion sizes.

STUDY DESIGN

One hundred sixty approximal noncavitated surfaces were radiographed using the Vistascan storage phosphor plate system. Original and enhanced (Fine, Caries1, and Caries2 filters) images were assessed by 6 observers who recorded the presence/absence of proximal carious lesions. Microscopy served to detect and measure true lesion size.

RESULTS

One hundred one surfaces were sound, and 59 had lesions. The sensitivity of the Fine filter images was significantly higher than the Caries1 and Caries2 filter modalities (P < .005). The original images did not differ significantly from the Caries1 and Caries2 modalities. Both the Fine filter and the original images had a significantly higher specificity than the Caries2 modality (P < .005). The overall accuracies for the Fine filter and the original images were significantly higher than for the Caries2 modality (P < .005). The Fine filter images also had higher accuracy than the Caries1 modality (P < .05). The mutual sensitivity of the filters varied little with lesion size.

CONCLUSION

Fine filter may be a promising enhancement tool for images used for detection of shallow carious lesion, because it showed less observer variability. The task-specific enhancement filters, Caries1 and Caries2, were less accurate than the original and Fine filter images and cannot be recommended for detection of the lesion sizes included in this study.

Alternative erasing times of the DenOptix system plate: performance on the detection of proximal caries

OBJECTIVE

Our aim was to investigate the DenOptix system for detection of proximal caries using alternative erasing times for the photostimulable phosphor plates.

STUDY DESIGN

Human teeth were X-rayed with phosphor plates using different erasing times. Five observers evaluated the images for the presence of caries by scoring the proximal surfaces of each tooth. Mean pixel intensity analyses were also carried out. The erasing times were evaluated using receiver operating characteristic (ROC) curve analysis. Analysis of variance for differences among erasing times and pixel intensity values was performed, followed by linear regression analysis for correlation between these variables.

RESULTS

The mean ROC curve for each erasing time varied from 0.61 to 0.66 (not significant). Analysis of variance did not show any significant effect of the erasing times on the pixel intensity values of the images (P > .05).

CONCLUSION

Erasing times for DenOptix phosphor plates from 20 to 130 s were satisfactory for diagnosis of proximal caries, and their use is therefore feasible in clinical practice.

Digital radiographic imaging: is the dental practice ready?

BACKGROUND

Digital radiographic imaging is slowly, but surely, replacing film-based imaging. It has many advantages over traditional imaging, but the technology also has some drawbacks. The author presents an overview of the types of digital image receptors available, image enhancement software and the range of costs for the new technology. PRACTICE IMPLICATIONS. The expenses associated with converting to digital radiographic imaging are considerable. The purpose of this article is to provide the clinician with an overview of digital radiographic imaging technology so that he or she can be an informed consumer when evaluating the numerous digital systems in the marketplace.

Perceptibility curve test for digital radiographs before and after application of various image processing algorithms

OBJECTIVES

The aim of the present study was to compare digital radiographs before and after the application of various digital image enhancement methods by means of a perceptibility curve (PC) test.

METHODS

Radiographs of an aluminium test object containing holes with sizes ranging from 0.5 mm to 1.5 mm were exposed at nine time settings, ranging from 0.05 s to 3.2 s using Digora (Soredex, Helsinki, Finland) storage phosphor system. Original digital images were then enhanced and displayed in colour-coded, contrast- and brightness-enhanced, histogram-equalized and negative modes. Thus, five series of images were created. Eight observers independently evaluated all of the images in the same random order and under the same viewing conditions. The object detail with the lowest perceptible contrast was recorded for each observer and each image type. Modified PCs were plotted according to the mean observer data. Repeated-measures ANOVA was used to compare the PCs and the perceptible number of details in original and enhanced images for each exposure (P=0.05). The Bonferroni/Dunn test was used to identify statistical differences among the groups.

RESULTS

Histogram equalization showed the maximum number of perceptible details in both the lowest optimal exposure time and the widest exposure range (P<0.05). Significant differences were present in low and middle exposure ranges of images enhanced with various modalities (P<0.05), while no difference was obtained among the number of details for different enhancements in the high exposure range (P>0.05).

CONCLUSION

Histogram-equalization enhancement of Digora images demonstrated subjectively superior image quality, followed by contrast and brightness enhancement.

Risk factors for a false positive test outcome in diagnosis of caries in approximal surfaces: impact of radiographic modality and observer characteristics

The aim of this study was to evaluate the impact of image modality and observer characteristics on radiographic caries diagnosis with particular emphasis on the false positive (FP) test outcome. Eighty human teeth were radiographed by eight digital modalities and film. Eight observers scored caries lesions in each approximal tooth surface. Their use of image enhancement and time spent in assessing an image was recorded. Microscopy served as validation for the presence/absence of a lesion. Of 160 approximal surfaces, 63% were sound and 37% had lesions. Multivariate logistic regression analysis was performed with the FP test outcome as the dependent variable and the radiographic modality and observer's gender, experience, employment, image enhancement, and time use as the independent variables. The significant risk factors for an FP test outcome were: (1) gender: males scored fewer FP than females (OR 0.26, CI 0.18-0.36); (2) experience: observers with no experience with digital radiography had a six times higher risk for an FP (OR 6.09, CI 4.9-8.1); (3) radiographic modality: two of the digital systems had a significantly lower risk for an FP [a phosphor plate system (OR 0.55, CI 0.37-0.73) and a sensor-based system (OR 0.65, CI 0.46-0.93)] and one phosphor plate system had a higher risk (OR 1.57, CI 1.17-2.13) than film; (4) time use: more time was spent when scoring an FP (OR 1.05, CI 1.03-1.05) than for any other test outcome. Type of employment and the use of image enhancement had no significant impact on the risk for an FP diagnosis.

Detection of Proximal Caries in vitro Using Standard and Task-Specific Enhanced Images from a Storage Phosphor Plate System

Eight dentists evaluated 72 proximal surfaces of premolars with respect to all caries lesions and to lesions into dentine in digital images from a storage phosphor plate system. The images were processed in four different ways: (1) the default algorithm of the standard imaging system (DF); (2) an algorithm correcting for attenuation and visual response (AV), and two proprietary caries-specific enhancement algorithms (3) K1, and (4) K2, respectively. The lesions were validated by histological examination. There were no significant differences in the areas under the receiver operating characteristic curves between differently processed radiographs for the categories of all caries lesions and caries into dentine.

A review of dentists' use of digital radiography and caries diagnosis with digital systems

OBJECTIVES

To investigate the evidence for (1) dentists' use of digital radiography and (2) the outcome of caries diagnosis with digital systems.

METHODS

A literature search with the software search package PubMed was used to get internet-based access to Medline through the website www.ncbi.nlm.nih.gov/pubmed. The search was limited to the years 1999-2005 since most papers dealing with the diagnostic value of digital radiography systems published before 1999 will hold little interest for today's users due to changes in the systems. The search strategies resulted in 123 articles (Table 1, #4 and #5). Original research articles (not reviews) were selected by the following inclusion criteria: (1) questionnaire studies on the use of direct digital intraoral radiography systems (not digitized film), (2) studies which used human teeth and natural caries lesions, and further in laboratory studies, the sectioned tooth was the gold standard for validating the presence or depth of a lesion.

RESULTS

The search resulted in 42 articles fulfilling the above criteria, which could be grouped into three types of studies: (a) questionnaire studies, (b) clinical (in vivo) studies, and (c) laboratory (in vitro) studies. Nine questionnaire studies, five clinical studies and 28 laboratory studies were found. These studies and their results are summarized in Tables 2-5.

CONCLUSIONS

The number of studies was limited, and some of the digital systems were evaluated in only one or two studies. A conclusive judgment may therefore not be possible for the majority of the digital systems selected for this review. There is a continuous need for the evaluation of new digital intraoral radiography systems that appear on the market, first and foremost for their image quality and diagnostic accuracy, but certainly also for their performance in the clinic, a clear deficiency observed after the literature search for the present review.

Influence of spatial resolution and bit depth on detection of small caries lesions with digital receptors

OBJECTIVE

The aim of this study was to compare the caries diagnostic accuracy among intraoral digital receptor systems that provide images in more than one resolution and bit depth.

STUDY DESIGN

Eighty noncavitated extracted human teeth were radiographed and analyzed by the following digital systems: Digora Optime (Soredex) 8-bit high- and 8-bit super-resolution, VistaScan (Dürr) 8-bit high- and 8-bit max-resolution and 16-bit high- and 16-bit max-resolution, Dixi2 (Planmeca) 8-bit and 12-bit depths. Insight Film was included as a reference. Six observers scored caries lesions in each tooth surface. The teeth were sectioned and microscopy served as validation. Two-way analysis of variance tested differences in sensitivity, specificity, positive and negative predictive values, and overall accuracy (true positives + true negatives) between the modalities.

RESULTS

There were 160 approximal surfaces examined, and histology showed 63% sound, 31% enamel, and 6% dentinal lesions. Sensitivity was not significantly different within each brand except for VistaScan 8-bit max-resolution, which showed higher sensitivity than the 2 VistaScan high-resolution modalities (P = .003). Digora super-resolution showed higher sensitivity than almost all the other modalities (P < .02), but had lower specificity than all other systems, among which there were no differences. Overall accuracy was significantly lower for Digora high-resolution than for Dixi2 and VistaScan max-resolution modalities (P < .03); there were no differences among the other modalities.

CONCLUSION

Caries diagnostic accuracy seems to be little influenced by an increase in spatial resolution and bit depth from 8-bit to 12- or 16-bit within digital radiographic system brands. Between systems, the Digora super-resolution images provided higher sensitivity, but lower specificity than most of the other receptors.

Accuracy of extraoral tuned aperture computed tomography (TACT) for proximal caries detection

OBJECTIVES

The objectives of this study were to compare the difference in the accuracy of proximal caries detection by extraoral tuned aperture computed tomography (TACT), intraoral TACT, and film radiographs.

STUDY DESIGN

Eighty proximal surfaces of 40 extracted human maxillary teeth were used. A digital sensor was the image receptor for TACT. Film radiographs were acquired using Insight film. Nine basis images were acquired to reconstruct TACT slices. Seven observers scored the presence or absence of proximal caries using the 3 imaging modalities. The true presence of caries and its depth were determined using the sectional images obtained by micro CT. Among the image modalities and observers, possible differences in the area under the receiver operating characteristic curve were assessed by analysis of variance (ANOVA).

RESULTS

ANOVA indicated no statistically significant differences between observers (P = .845), modalities (P = .657), and observer-modality combinations (P = .593).

CONCLUSION

Within the limited range of this study, extraoral TACT was not statistically different from intraoral TACT or film radiographs for proximal caries detection. This suggests that extraoral TACT may have some clinical utility for caries diagnosis and that further study may be warranted.

Sliding window adaptive histogram equalization of intraoral radiographs: effect on image quality

OBJECTIVES

To investigate whether contrast enhancement by non-interactive, sliding window adaptive histogram equalization (SWAHE) can enhance the image quality of intraoral radiographs in the dental clinic.

METHODS

Three dentists read 22 periapical and 12 bitewing storage phosphor (SP) radiographs. For the periapical readings they graded the quality of the examination with regard to visually locating the root apex. For the bitewing readings they registered all occurrences of approximal caries on a confidence scale. Each reading was first done on an unprocessed radiograph ("single-view"), and then re-done with the image processed with SWAHE displayed beside the unprocessed version ("twin-view"). The processing parameters for SWAHE were the same for all the images.

RESULTS

For the periapical examinations, twin-view was judged to raise the image quality for 52% of those cases where the single-view quality was below the maximum. For the bitewing radiographs, there was a change of caries classification (both positive and negative) with twin-view in 19% of the cases, but with only a 3% net increase in the total number of caries registrations. For both examinations interobserver variance was unaffected.

CONCLUSIONS

Non-interactive SWAHE applied to dental SP radiographs produces a supplemental contrast enhanced image which in twin-view reading improves the image quality of periapical examinations. SWAHE also affects caries diagnosis of bitewing images, and further study using a gold standard is warranted.

Diagnostic accuracy of two software modalities for detection of caries lesions in digital radiographs from four dental systems

OBJECTIVE

To compare the caries diagnostic accuracy of two software modalities used in the assessment of digital radiographs obtained with four different dental systems, and to evaluate whether the software used for image assessment influenced the mutual comparison of those four dental systems relating to their caries diagnostic accuracy.

METHODS

Under in vitro and standardized conditions 122 teeth (with 228 unrestored approximal and 99 occlusal surfaces) were radiographed in blocks of 3 test teeth and 2 non-test teeth using two storage phosphor plate systems: Digora (Soredex, Helsinki, Finland) and DenOptix (Gendex, Dentsply, Milan, Italy) and two charge coupled device (CCD)-based sensor systems: Dixi (Planmeca, Helsinki, Finland) and Sidexis (Sirona, Bensheim, Germany). The images were displayed and examined in two software modalities: their own dedicated software and a general software. Three observers examined all images for the presence of approximal enamel and dentine and occlusal dentine caries lesions using a 5-point confidence scale. The true presence of caries was validated by ground section histology. The diagnostic accuracy of the software modalities was expressed as ROC curve areas (A(z)) and differences between modalities were tested by paired t-test. Comparison of systems was analysed by post hoc t-test.

RESULTS

Results of approximal and occlusal surfaces assessed together revealed nearly identical mean A(z) with the two software modalities on images obtained with the Digora (A(z)=0.71) and DenOptix (A(z)=0.72) systems. On Dixi images the mean A(z) was 0.75 using the system's own software and 0.73 using the general software. On Sidexis images the corresponding mean A(z)s were 0.79 and 0.75. None of those differences were significant.

CONCLUSION

No significant difference in caries diagnostic accuracy was found between two software modalities used for examination of digital radiographs obtained with four different digital systems, and the software modality did not influence the mutual rank of the four systems relating to their diagnostic accuracy.