Distortion of digital panoramic radiographs used for implant site assessment

Advancements in artificial intelligence algorithms for dental implant identification: A systematic review with meta-analysis

STATEMENT OF PROBLEM

The evidence regarding the application of artificial intelligence (AI) in identifying dental implant systems is currently inconclusive. The available studies present varying results and methodologies, making it difficult to draw definitive conclusions.

PURPOSE

The purpose of this systematic review with meta-analysis was to comprehensively analyze and evaluate articles that investigate the application of AI in identifying and classifying dental implant systems.

MATERIAL AND METHODS

An electronic systematic review was conducted across 3 databases: MEDLINE/PubMed, Cochrane, and Scopus. Additionally, a manual search was performed. The inclusion criteria consisted of peer-reviewed studies investigating the accuracy of AI-based diagnostic tools on dental radiographs for identifying and classifying dental implant systems and comparing the results with those obtained by expert judges using manual techniques-the search strategy encompassed articles published until September 2023. The Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool was used to assess the quality of included articles.

RESULTS

Twenty-two eligible articles were included in this review. These articles described the use of AI in detecting dental implants through conventional radiographs. The pooled data showed that dental implant identification had an overall accuracy of 92.56% (range 90.49% to 94.63%). Eleven studies showed a low risk of bias, 6 demonstrated some concern risk, and 5 showed a high risk of bias.

CONCLUSIONS

AI models using panoramic and periapical radiographs can accurately identify and categorize dental implant systems. However, additional well-conducted research is recommended to identify the most common implant systems.

Classification of dental implant systems using cloud-based deep learning algorithm: an experimental study

BACKGROUND

This study aimed to evaluate the accuracy and clinical usability of implant system classification using automated machine learning on a Google Cloud platform.

METHODS

Four dental implant systems were selected: Osstem TSIII, Osstem USII, Biomet 3i Os-seotite External, and Dentsply Sirona Xive. A total of 4,800 periapical radiographs (1,200 for each implant system) were collected and labeled based on electronic medical records. Regions of interest were manually cropped to 400×800 pixels, and all images were uploaded to Google Cloud storage. Approximately 80% of the images were used for training, 10% for validation, and 10% for testing. Google automated machine learning (AutoML) Vision automatically executed a neural architecture search technology to apply an appropriate algorithm to the uploaded data. A single-label image classification model was trained using AutoML. The performance of the mod-el was evaluated in terms of accuracy, precision, recall, specificity, and F1 score.

RESULTS

The accuracy, precision, recall, specificity, and F1 score of the AutoML Vision model were 0.981, 0.963, 0.961, 0.985, and 0.962, respectively. Osstem TSIII had an accuracy of 100%. Osstem USII and 3i Osseotite External were most often confused in the confusion matrix.

CONCLUSION

Deep learning-based AutoML on a cloud platform showed high accuracy in the classification of dental implant systems as a fine-tuned convolutional neural network. Higher-quality images from various implant systems will be required to improve the performance and clinical usability of the model.

Detection of pulpal calcifications on bite-wing radiographs using deep learning

OBJECTIVES

Pulpal calcifications are discrete hard calcified masses of varying sizes in the dental pulp cavity. This study is aimed at measuring the performance of the YOLOv4 deep learning algorithm to automatically determine whether there is calcification in the pulp chambers in bite-wing radiographs.

MATERIALS AND METHODS

In this study, 2000 bite-wing radiographs were collected from the faculty database. The oral radiologists labeled the pulp chambers on the radiographs as "Present" and "Absent" according to whether there was calcification. The data were randomly divided into 80% training, 10% validation, and 10% testing. The weight file for pulpal calcification was obtained by training the YOLOv4 algorithm with the transfer learning method. Using the weights obtained, pulp chambers and calcifications were automatically detected on the test radiographs that the algorithm had never seen. Two oral radiologists evaluated the test results, and performance criteria were calculated.

RESULTS

The results obtained on the test data were evaluated in two stages: detection of pulp chambers and detection of pulpal calcification. The detection performance of pulp chambers was as follows: recall 86.98%, precision 98.94%, F1-score 91.60%, and accuracy 86.18%. Pulpal calcification "Absent" and "Present" detection performance was as follows: recall 86.39%, precision 85.23%, specificity 97.94%, F1-score 85.49%, and accuracy 96.54%.

CONCLUSION

The YOLOv4 algorithm trained with bite-wing radiographs detected pulp chambers and calcification with high success rates.

CLINICAL RELEVANCE

Automatic detection of pulpal calcifications with deep learning will be used in clinical practice as a decision support system with high accuracy rates in diagnosing dentists.

Influence of Exposure Parameters and Implant Position in Peri-Implant Bone Assessment in CBCT Images: An In Vitro Study

The aim of this study was to assess the impact of dimensional distortion and its changes with modification of exposure setting parameters on the measurements of peri-implant bone margin. Ten titanium dental implants (InKone Primo, Global D, Paris, France) were placed in two prepared bovine ribs. Two bone models and an implant-with-transfer model were scanned with 3shape E4 (3shape, Copenhagen, Denmark) laboratory scanner. Cone beam computed tomography (CBCT) images of two bone models were taken with different values of voltage (60, 70, 80, 90 kV), tube current (4, 10 mA) and voxel size (200, 300 µm). All the data were superimposed using planning software, and the measurements of buccal bone thickness in two selected regions were performed both using CBCT and scan cross-sections. The mean squared error (MSE) being the squared differences between measurements was used in the accuracy assessment of the CBCT device. A one-way ANOVA revealed significant differences between voltage and MSE (p = 0.044), as well as implant position and MSE (p = 0.005). The distortions of measurements depend on bone margin thickness, and the higher the distance to measure, the higher the error. Accurate measurements of buccal bone thickness (MSE below 0.25) were achieved with voltage values of 70, 80, and 90 kV.

A retrospective institutional study of human age determination by evaluating the pulp length and width ratio of the maxillary lateral incisor on panoramic radiographs in Indonesian subjects

Purpose

The pulp length to width (PL/W) ratio of the maxillary lateral incisor can be used as an age determination method. This study aimed to investigate the correlation between the PL/W ratio of the maxillary lateral incisor on panoramic radiographs and human chronological age in Indonesian subjects.

Materials and Methods

This study analyzed with 134 maxillary lateral incisors on 113 panoramic radiographs from patients who visited the Oral and Maxillofacial Radiology Unit of Dental Hospital Universitas Padjadjaran, Bandung, Jawa Barat, Indonesia, from 2013 to 2018 (age range: between 11 to 60 years). The pulp length was measured from the pulp chamber roof to the apical foramen, and the pulp width was measured on the cervical area of the cementoenamel junction in millimeters using Fiji ImageJ open-source software. Simple linear regression (in SPSS) was used to analyze the results. The reliability of the observers was evaluated.

Results

The PL/W ratio of the maxillary lateral incisor was significantly correlated with chronological age (P<0.01). No statistically significant difference was found in the PL/W ratio between the left and right maxillary lateral incisors (P=0.333). There was a very strong correlation (r=0.939) between the PL/W ratio of the maxillary lateral incisor and human chronological age, with the following formula: age= -3.057+1.875×PL/W ratio (R²=0.882, standard error of estimate: 4.659).

Conclusion

The PL/W ratio of the maxillary lateral incisor on panoramic radiograph can be used for age determination in Indonesian subjects.

Evaluation of the relation between the pulp stones and direct restorations using cone beam computed tomography in a Turkish subpopulation

Objectives

This study aimed to assess the presence of pulp stones through an examination of cone beam computed tomography images and correlate their prevalence with age, sex, dental arch and side, tooth type, and restoration type and depth.

Materials and Methods

Cone beam computed tomography images obtained from 673 patients and archival data on 11,494 teeth were evaluated. The associations of pulp stones with age, sex, dental arch and side, tooth type, and restoration type and depth were noted. All the measurements were subjected to a χ² test and one sample χ² test (p < 0.05).

Results

In the study group, 163 (24.2%) patients and 379 (3.3%) teeth had at least one pulp stone. The pulp stone frequency in those aged 30–39 years was significantly greater than in those aged 18–29 and ≥ 60 years, and the frequency was higher in females than in males (p < 0.05). The highest prevalence of pulp stones was found in maxillary dental arches and molar teeth (p < 0.05). Pulp stones were significantly more common in medium-depth restorations (p < 0.05).

Conclusions

Maxillary molar teeth, medium-depth restorations, individuals aged 30–39 years and females had a greater percentage of pulp stones.

Incidence and distribution of pulp stones among Southern Saudi Arabian sub-population

Purpose:

To assess the incidence of pulp stones in molar and premolar teeth of Southern Saudi Arabian adult sub-population.

Materials and Methods:

Six hundred patient files were randomly selected from the database (records) of the College of Dentistry outpatient department screened by Oral Medicine and Radiology division. Orthopantomogram and bitewing radiographs using radiovisiographs of first and second maxillary and mandibular molars and premolars were interpreted by three examiners. Pulp stones were identified as definite radiopaque masses and scored as present or absent. Data were subjected to statistical analysis using SPSS version 19.

Results:

Out of total 600 patients, pulp stones were found in 88 (14.7%) patients. Females showed statistically significant higher presence of pulp stones in mandibular molars than males. The patients belonging to middle age group (31–40 years) showed statistically significant higher presence of pulp stones than the other two groups in all first and second molar teeth in both arches. The presence of pulp stones was observed to be higher in maxillary arch than mandibular arch with left side having more frequency than the right side of the arch.

Conclusion:

The prevalence of pulp stones in the study population was found to be 14.7%, and it was higher in molars than premolars. It is quite significant to know the frequency distribution of pulp stones especially for an endodontist or a dentist to plan the root canal treatment of such teeth accordingly.

Efficacy of deep convolutional neural network algorithm for the identification and classification of dental implant systems, using panoramic and periapical radiographs: A pilot study

Convolutional neural networks (CNNs), a particular type of deep learning architecture, are positioned to become one of the most transformative technologies for medical applications. The aim of the current study was to evaluate the efficacy of deep CNN algorithm for the identification and classification of dental implant systems.A total of 5390 panoramic and 5380 periapical radiographic images from 3 types of dental implant systems, with similar shape and internal conical connection, were randomly divided into training and validation dataset (80%) and a test dataset (20%). We performed image preprocessing and transfer learning techniques, based on fine-tuned and pre-trained deep CNN architecture (GoogLeNet Inception-v3). The test dataset was used to assess the accuracy, sensitivity, specificity, receiver operating characteristic curve, area under the receiver operating characteristic curve (AUC), and confusion matrix compared between deep CNN and periodontal specialist.We found that the deep CNN architecture (AUC = 0.971, 95% confidence interval 0.963-0.978) and board-certified periodontist (AUC = 0.925, 95% confidence interval 0.913-0.935) showed reliable classification accuracies.This study demonstrated that deep CNN architecture is useful for the identification and classification of dental implant systems using panoramic and periapical radiographic images.

Image quality assessment of pre-processed and post-processed digital panoramic radiographs in paediatric patients with mixed dentition

Purpose

To determine the impact of an image processing technique on diagnostic accuracy of digital panoramic radiographs for the assessment of anatomical structures in paediatric patients with mixed dentition.

Materials and Methods

The study consisted of 50 digital panoramic radiographs of children aged from 6 to 12 years, which were later on processed using a dedicated image processing method. A modified clinical image quality evaluation chart was used to evaluate the diagnostic accuracy of anatomical structures in maxillary and mandibular anterior and maxillary premolar region of processed images.

Results

A statistically significant difference was observed between pre and post-processed evaluation of anatomical structures (P<0.05) in the maxillary and mandibular anterior region. The anterior region was found to be more accurate in post-processed images. No significant difference was observed in the maxillary premolar region (P>0.05). The Inter-observer and intra-observer reliability of both pre and post processed images were excellent (>0.82) for anterior region and good (>0.63) for premolar region.

Conclusion

The application of image processing technique in digital panoramic radiography can be considered a reliable method for improving the quality of anatomical structures in paediatric patients with mixed dentition.

Evaluation of the relationship between sleep bruxism and pulpal calcifications in young women: A clinico-radiological study

PURPOSE

This study was performed to investigate the relationship between sleep bruxism (SB) and pulpal calcifications in young women.

MATERIALS AND METHODS

A total of 100 female participants between 20 and 31 years of age who were referred to our radiology clinic for a dental check-up, including 59 SB and 41 non-SB patients, were sampled for the analysis. SB was diagnosed based on the American Academy of Sleep Medicine criteria. All teeth were evaluated on digital panoramic radiographs to detect pulpal calcifications, except third molars, teeth with root canal treatment, and teeth with root resorption. Binary logistic regression analysis was used to determine the risk factors for pulpal calcifications. The Spearman correlation coefficient was applied and the Pearson chi-square test was used for categorical variables. To test intra-examiner reproducibility, Cohen kappa analysis was applied. P values <.05 were considered to indicate statistical significance.

RESULTS

A total of 2800 teeth were evaluated (1652 teeth from SB patients and 1148 from non-SB patients), and 61% of patients had at least 1 dental pulpal calcification. No statistically significant relationship was found between SB and pulpal calcifications (P>0.05). In SB patients, the total number of pulpal calcifications was 129, while in non-SB patients, it was 84. Binary logistic analysis showed that SB was not a risk factor for the presence of pulpal calcifications (odds ratio, 1.19; 95% CI, 0.52-2.69, P>.05).

CONCLUSION

No relationship was found between SB and pulpal calcifications.

A comparative study of cone-beam computed tomography and digital panoramic radiography for detecting pulp stones

PURPOSE

The aim of this study was to compare cone-beam computed tomography (CBCT) and digital panoramic radiography (DPR) for the detection of pulp stones.

MATERIALS AND METHODS

DPR and CBCT images of 202 patients were randomly selected from the database of our department. All teeth were evaluated in sagittal, axial, and coronal sections in CBCT images. The systemic condition of patients, the presence of pulp stones, the location of the tooth, the group of teeth, and the presence and depth of caries and restorations were recorded. The presence of pulp stones in molar teeth was compared between DPR and CBCT images.

RESULTS

Pulp stones were identified in 105 (52.0%) of the 202 subjects and in 434 (7.7%) of the 5,656 teeth examined. The prevalence of pulp stones was similar between the sexes and across various tooth locations and groups of teeth (P>.05). A positive correlation was observed between age and the number of pulp stones (ρ=0.277, P<.01). Pulp stones were found significantly more often in restored or carious teeth (P<.001). CBCT and DPR showed a significant difference in the detection of pulp stones (P<.001), which were seen more often on DPR than on CBCT.

CONCLUSION

DPR, as a 2D imaging system, has inherent limitations leading to the misinterpretation of pulp stones. Restored and carious teeth should be carefully examined for the presence of pulp stones. CBCT imaging is recommended for a definitive assessment in cases where there is a suspicion of a pulp stone on DPR.