• Restorative Dentistry and Endodontics
• /
• v.46 no.2
• /
• pp.23.1-23.11
• /
• 2021

Objectives: This study assessed the reliability of digital color measurements using images of resin composite specimens captured with a cellphone. Materials and Methods: The reference color of cylindrical specimens built-up with the use of resin composite (shades A1, A2, A3, and A4) was measured with a portable spectrophotometer (CIELab). Images of the specimens were obtained individually or pairwise (compared shades in the same photograph) under standardized parameters. The color of the specimens was measured in the images using RGB system and converted to CIELab system using image processing software. Whiteness index (WI_D) and color differences (ΔE₀₀) were calculated for each color measurement method. For the cellphone, the ΔE₀₀ was calculated between the pairs of shades in separate images and in the same image. Data were analyzed using 2-way repeated-measures analysis of variance (α = 0.05). Linear regression models were used to predict the reference ΔE₀₀ values of those calculated using color measured in the images. Results: Images captured with the cellphone resulted in different WI_D values from the spectrophotometer only for shades A3 and A4. No difference to the reference ΔE₀₀ was observed when individual images were used. In general, a similar ranking of ΔE₀₀ among resin composite shades was observed for all methods. Stronger correlation coefficients with the reference ΔE₀₀ were observed using individual than pairwise images. Conclusions: This study showed that the use of cellphone images to measure the color difference seems to be a feasible alternative providing outcomes similar to those obtained with the spectrophotometer.

• Imaging Science in Dentistry
• /
• v.32 no.4
• /
• pp.187-194
• /
• 2002

Purpose: To evaluate the performance of the LOGICON Caries Detector using RVG-4 and RVG-ui sensors, by comparing results of each detector to the results of clinical and histological examinations. Materials and Methods : Pairs of extracted teeth were radiographed, and a total of 57 proximal surfaces, which included both carious and non-carious situations, were analyzed. The RVG-4 produced 8-bit images, while the RVG-ui unit produced 12-bit images, which were taken in the high sensitivity mode. The images produced by the LOGICON were evaluated by a trained observer using both automated and manual caries detection software modes. Ground sections of the teeth established the actual absence or existence of caries. Results: LOGIC ON-aided caries detection and depth discrimination of the RVG-4 and RVG-ui sensors were equally inconsistent irrespective of whether the LOGIC ON software was set to the automated or manual mode. Sensitivity ranged from 50% to 57% for caries penetration of the enamel-dentin junction. Conclusion: Care needs to be taken when using LOGIC ON in conjunction with RVG images as an adjunct for treatment planning dental caries. Even when applied by a trained observer, substantial discrepancies exist between the results of the LOGIC ON software-guided evalutations using RVG images and histologic examination.

• Imaging Science in Dentistry
• /
• v.40 no.3
• /
• pp.103-107
• /
• 2010

Purpose : This study was performed to examine the effects of image filter on observer performance by counting the number of holes at each wedge step on a radiographic image. Materials and Methods : An aluminum step wedge with 11 steps ranged in thickness from 1.5 mm to 16.5 mm in 1.5 mm increments was fabricated for this study. Each step had 10 notched holes with 1.0 mm diameter on the bottom of the step wedge which were ranged in depths from 0.1 mm to 1.0 mm in 0.1 mm increments. Digital radiographic raw images of the aluminum step wedge were acquired by using CCD intraoral sensor. The images were processed using several types of noise reduction filters and kernel sizes. Three observers counted the number of holes which could be discriminated on each step. The data were analyzed by ANOVA. Results : The number of holes at each step was decreased as the thickness of step was increased. The number of holes at each step on the raw images was significantly higher than that on the processed images. The number of holes was different according to the types and kernel sizes of the image filters. Conclusions : The types and kernel sizes of image filters on observer performance were important, therefore, they should be standardized for commercial digital imaging systems.

• Imaging Science in Dentistry
• /
• v.47 no.4
• /
• pp.261-267
• /
• 2017

Purpose: Analyzing different patterns of maxillary sinus septa in cone-beam computed tomography (CBCT) images and predicting maxillary sinus membrane perforations. Materials and Methods: In this cross-sectional study, CBCT images of 222 patients ranging from 20 to 81 years old were evaluated. One hundred fifty-two patients (93 females and 59 males) who had maxillary sinus septa in axial views were included in this study. Cross-sectional images were used to determine classifications of sinus septa and the risk of membrane perforation using a method modified from Al-Faraje et al. Variables of sex, age, and dental status were considered. Chi-squared and Kruskal-Wallis tests were used for data analysis(P<.05). Results: In this study, 265 maxillary sinus septal patterns were found. The mean age of the patients was $44.1{\pm}14.7$ years old. The Class I and VII-div II patterns had the greatest and least prevalence, respectively. Furthermore, there was a significant relationship between the location of septa and the frequency of membrane perforation risk (P<.05). In this study, the relationship of different patterns of septa with dental status did not differ significantly (P>0.05). Conclusion: A higher prevalence of moderate risk of membrane perforation in the molar region relative to the premolar region was observed. Furthermore, maxillary sinus septa occur most frequently in the molar region, demonstrating the importance of paying attention to this region during sinus lift surgery. This study did not show any relationship between tooth loss and the presence of septa.

• Imaging Science in Dentistry
• /
• v.49 no.3
• /
• pp.191-199
• /
• 2019

Purpose: To compare the diagnostic ability of undergraduate dental students to detect maxillary sinus abnormalities in panoramic radiographs(PR) and cone-beam computed tomography (CBCT). Materials and Methods: This was a retrospective study based on the evaluation of PR and CBCT images. A pilot study was conducted to determine the number of students eligible to participate in the study. The images were evaluated by 2 students, and 280 maxillary sinuses were assessed using the following categories: normal, mucosal thickening, sinus polyp, antral pseudocyst, nonspecific opacification, periostitis, antrolith, and antrolith associated with mucosal thickening. The reference standard was established by the consensus of 2 oral radiologists based on the CBCT images. The kappa test, receiver operating characteristic curves, and 1-way analysis of variance with the Tukey-Kramer post-hoc test were employed. Results: Intraobserver and interobserver reliability showed agreement ranging from substantial (0.809) to almost perfect (0.922). The agreement between the students' evaluations and the reference standard was reasonable (0.258) for PR and substantial(0.692) for CBCT. Comparisons of values of sensitivity, specificity, and accuracy showed that CBCT was significantly better(P<0.05). Conclusion: CBCT was better than PR for the detection of maxillary sinus abnormalities by dental students. However, CBCT should only be requested after a careful analysis of PR by students and more experienced professionals.

• Journal of Veterinary Science
• /
• v.21 no.5
• /
• pp.75.1-75.8
• /
• 2020

Background: Dental diseases are common in dogs and cats, and accurate measurements of dentoalveolar structure are important for planning of treatment. The information that the comparison computed tomography (CT) with dental radiography (DTR) is not yet reported in veterinary medicine. Objectives: The purpose of this study was to compare the DTR with CT of dentoalveolar structures in healthy dogs and cats, and to evaluate the CT images of 2 different slice thicknesses (0.5 and 1.0 mm). Methods: We included 6 dogs (2 Maltese and 1 Spitz, Beagle, Pomeranian, mixed, 1 to 8 years, 4 castrated males, and 2 spayed female) and 6 cats (6 domestic short hair, 8 months to 3 years, 4 castrated male, and 2 spayed female) in this study. We measured the pulp cavity to tooth width ratio (P/T ratio) and periodontal space of maxillary and mandibular canine teeth, maxillary fourth premolar, mandibular first molar, maxillary third premolar and mandibular fourth premolar. Results: P/T ratio and periodontal space in the overall dentition of both dogs and cats were smaller in DTR compared to CT. In addition, CT images at 1.0 mm slice thickness was generally measured to be greater than the images at 0.5 mm slice thickness. Conclusions: The results indicate that CT with thin slice thickness provides more accurate information on the dentoalveolar structures. Additional DTR, therefore, may not be required for evaluating dental structure in small-sized dogs and cats.

• Imaging Science in Dentistry
• /
• v.54 no.3
• /
• pp.257-263
• /
• 2024

Purpose: The purpose of this study was to classify mandibular molar furcation involvement (FI) in periapical radiographs using a deep learning algorithm. Materials and Methods: Full mouth series taken at East Carolina University School of Dental Medicine from 2011-2023 were screened. Diagnostic-quality mandibular premolar and molar periapical radiographs with healthy or FI mandibular molars were included. The radiographs were cropped into individual molar images, annotated as "healthy" or "FI," and divided into training, validation, and testing datasets. The images were preprocessed by PyTorch transformations. ResNet-18, a convolutional neural network model, was refined using the PyTorch deep learning framework for the specific imaging classification task. CrossEntropyLoss and the AdamW optimizer were employed for loss function training and optimizing the learning rate, respectively. The images were loaded by PyTorch DataLoader for efficiency. The performance of ResNet-18 algorithm was evaluated with multiple metrics, including training and validation losses, confusion matrix, accuracy, sensitivity, specificity, the receiver operating characteristic (ROC) curve, and the area under the ROC curve. Results: After adequate training, ResNet-18 classified healthy vs. FI molars in the testing set with an accuracy of 96.47%, indicating its suitability for image classification. Conclusion: The deep learning algorithm developed in this study was shown to be promising for classifying mandibular molar FI. It could serve as a valuable supplemental tool for detecting and managing periodontal diseases.

• Imaging Science in Dentistry
• /
• v.36 no.3
• /
• pp.123-129
• /
• 2006

The use of computed tomography for dental procedures has increased recently. Cone beam computed tomography (CBCT) systems have been designed for imaging hard tissues of the dentomaxillofacial region. CBCT is capable of providing high resolution in images of high diagnostic quality. This technology allows for 3-dimensional representation of the dentomaxillofacial skeleton with minimal distortion, but at lower equipment cost, simpler image acquisition and lower patient dose. Because this technology produces images with isotropic sub-millimeter spatial resolution, it is ideally suited for dedicated dentomaxillofacial imaging. In this paper, we provide a brief overview of cone beam scanning technology and compare it with the fan beam scanning used in conventional CT and the basic principles of currently available CBCT systems.

• Imaging Science in Dentistry
• /
• v.45 no.2
• /
• pp.81-87
• /
• 2015

Purpose: Contrast, sharpness, enhancement, and density can be changed in digital systems. The important question is to what extent the changes in these variables affect the accuracy of caries detection. Materials and Methods: Forty eight extracted human posterior teeth with healthy or proximal caries surfaces were imaged using a photostimulable phosphor (PSP) sensor. All original images were processed using a six-step method: (1) applying "Sharpening 2" and "Noise Reduction" processing options to the original images; (2) applying the "Magnification 1:3" option to the image obtained in the first step; (3) enhancing the original images by using the "Diagonal/"option; (4) reviewing the changes brought about by the third step of image processing and then, applying "Magnification 1:3"; (5) applying "Sharpening UM" to the original images; and (6) analyzing the changes brought about by the fifth step of image processing, and finally, applying "Magnification 1:3." Three observers evaluated the images. The tooth sections were evaluated histologically as the gold standard. The diagnostic accuracy of the observers was compared using a chi-squared test. Results: The accuracy levels irrespective of the image processing method ranged from weak (18.8%) to intermediate (54.2%), but the highest accuracy was achieved at the sixth image processing step. The overall diagnostic accuracy level showed a statistically significant difference (p=0.0001). Conclusion: This study shows that the application of "Sharpening UM" along with the "Magnification 1:3" processing option improved the diagnostic accuracy and the observer agreement more effectively than the other processing procedures.

• Journal of dental hygiene science
• /
• v.20 no.4
• /
• pp.206-212
• /
• 2020

Background: A Teachable Machine is a kind of machine learning web-based tool for general persons. In this paper, the feasibility of Google's Teachable Machine (ver. 2.0) was studied in the diagnosis of the tooth-marked tongue. Methods: For machine learning of tooth-marked tongue diagnosis, a total of 1,250 tongue images were used on Kaggle's web site. Ninety percent of the images were used for the training data set, and the remaining 10% were used for the test data set. Using Google's Teachable Machine (ver. 2.0), machine learning was performed using separated images. To optimize the machine learning parameters, I measured the diagnosis accuracies according to the value of epoch, batch size, and learning rate. After hyper-parameter tuning, the ROC (receiver operating characteristic) analysis method determined the sensitivity (true positive rate, TPR) and specificity (false positive rate, FPR) of the machine learning model to diagnose the tooth-marked tongue. Results: To evaluate the usefulness of the Teachable Machine in clinical application, I used 634 tooth-marked tongue images and 491 no-marked tongue images for machine learning. When the epoch, batch size, and learning rate as hyper-parameters were 75, 0.0001, and 128, respectively, the accuracy of the tooth-marked tongue's diagnosis was best. The accuracies for the tooth-marked tongue and the no-marked tongue were 92.1% and 72.6%, respectively. And, the sensitivity (TPR) and specificity (FPR) were 0.92 and 0.28, respectively. Conclusion: These results are more accurate than Li's experimental results calculated with convolution neural network. Google's Teachable Machines show good performance by hyper-parameters tuning in the diagnosis of the tooth-marked tongue. We confirmed that the tool is useful for several clinical applications.