Prevalence of calcifications in soft tissues visible on a dental pantomogram: A retrospective analysis

Differential Diagnosis of OKC and SBC on Panoramic Radiographs: Leveraging Deep Learning Algorithms

This study aims to determine whether it can distinguish odontogenic keratocyst (OKC) and simple bone cyst (SBC) based solely on preoperative panoramic radiographs through a deep learning algorithm. (1) Methods: We conducted a retrospective analysis of patient data from January 2018 to December 2022 at Pusan National University Dental Hospital. This study included 63 cases of OKC confirmed by histological examination after surgical excision and 125 cases of SBC that underwent surgical curettage. All panoramic radiographs were obtained utilizing the Proline XC system (Planmeca Co., Helsinki, Finland), which already had diagnostic data on them. The panoramic images were cut into 299 × 299 cropped sizes and divided into 80% training and 20% validation data sets for 5-fold cross-validation. Inception-ResNet-V2 system was adopted to train for OKC and SBC discrimination. (2) Results: The classification network for diagnostic performance evaluation achieved 0.829 accuracy, 0.800 precision, 0.615 recall, and a 0.695 F1 score. (4) Conclusions: The deep learning algorithm demonstrated notable accuracy in distinguishing OKC from SBC, facilitated by CAM visualization. This progress is expected to become an essential resource for clinicians, improving diagnostic and treatment outcomes.

Prevalence and Radiographic Features of Head and Neck Soft Tissue Calcifications on Digital Panoramic Radiographs: A Retrospective Study

Background In this study, we aimed to determine the prevalence and radiographic features of incidental head and neck soft tissue calcifications (STCs) on panoramic imagesand assess their clinical significance. Methodology Following well-established training and calibration procedures, 9,553 digital panoramic radiographs (DPRs) taken between January 1, 2021, and January 31, 22, were retrospectively evaluated. Only obvious calcifications and clear differential diagnoses were considered. The presence, type, side (i.e., unilateral or bilateral), number (single or multiple), and the presence of different calcifications in the same individual were recorded. STCs were recorded according to age and gender. Data were analyzed using the chi-square test and Fisher's exact test using SPSS version 18.0 (IBM Corp., Armonk, NY, USA). Results Overall, 35.8% of the DPRs studied showed the presence of STCs, including ossified stylohyoid complex (OSHC) (10.3%), thyroid cartilage (9.8%), tonsillolith (9.2%), atherosclerotic plaques (5.8%), calcified triticeous cartilage (CTC) (5.1%), sialolith (1.9%), as well as intra-articular (1.3%) and other calcifications (0.1-0.8%), i.e., calcified lymph node, antrolith, rhinolith, phlebolith, and osteoma cutis. STCs were found to be more prevalent in middle-aged patients and in females. A significant relationship was identified between the presence of carotid artery calcification and calcified superior horn of thyroid cartilage (CSHTC), as well as between the presence of CSHTC and CTC. Calcifications were detected either bilaterally (n = 2,003) or unilaterally (n = 2,388); however, OSHC mostly showed bilateral calcifications (8.5%). Conclusions Panoramic radiographs of dental patients reveal the frequent occurrence of STCs in the head and neck region with differing radiographic features. Certain calcifications show gender and age differences. Accurate detection of STCs may guide the identification of potential underlying diseases and help initiate referral to the relevant multidisciplinary teams.

Prevalence of incidental findings in adult vs. adolescent patients in the course of orthodontic X-ray diagnostics

OBJECTIVE

Due to increasing numbers of adult patients presenting to orthodontic practices, an increase in incidental findings on diagnostic X‑rays, which are the cornerstone of orthodontic diagnostics, is expected. This raises the clinically relevant question of whether an age effect exists regarding prevalence, localisation and severity of incidental findings on orthodontic diagnostic X‑rays.

MATERIALS AND METHODS

The clinical, primarily retrospective study examined pathological incidental findings from 600 orthopantomograms (OPT) and lateral cephalogram (LC) images in two groups of orthodontic patients (group I: 150 children/adolescents, age 11.89 ± 2.47 years; group II: 150 adults, age 27.03 ± 10.42 years). Prevalence, localisation and severity of the findings were recorded based on a classification sheet. The assessment was done by three experienced examiners following a systematic approach along the nine locations: mandible, maxilla, dentition, paranasal sinuses, temporomandibular joint, cranial base, orbit, cervical spine, soft tissues.

RESULTS

In all, 1458 incidental findings were detected, with 66% of the findings having occurred away from the dentition. There was a significant age effect (p < 0.001) with respect to the prevalence of incidental findings (group II-adults 1026 findings-OPT: 566/LC 460 vs. group I-children/adolescents 432 findings-OPT: 221/LC 211). Regarding localisation, incidental findings in adults commonly occurred in the dentition, paranasal sinuses and mandibular regions. Furthermore, analysis of the LC images revealed significantly more incidental findings in the area of the cranial base and cervical spine in adults (p < 0.001, p = 0.003). Categorisation according to the severity of the incidental findings showed that 33% of the incidental findings needed further diagnostic investigation and possibly treatment by other specialities.

CONCLUSION

Diagnostic assessment using orthodontic diagnostic X‑rays results in a high prevalence of incidental findings away from the dentition. Particularly in adults, a large number of incidental findings outside the dental/alveolar region may be expected on orthodontic diagnostic X‑rays. Thus, a structured approach during diagnostic assessment is required to minimise the extent to which incidental findings of clinical relevance are overlooked.

Carotid Artery Calcifications on Panoramic Radiographs

OBJECTIVE

The objective of this study was to evaluate the prevalence of carotid artery calcifications (CACs) on panoramic radiographs of Polish dental patients.

METHODS

Panoramic radiographs made between 2005 and 2012 in 4000 patients born between 1922 and 1958 were accessed from the server of the Department of Radiology and analyzed for the presence or absence of CACs by a group of trained dentists.

RESULTS

The anatomical area covered by the analysis was visible in 2189 images (54.73%). Calcifications in the carotid arteries were found in 468 (21.68%) patients, including 327 (14.94%) unilaterally and 141 (6.44%) bilaterally. CACs were found in 284 (60.68%) women and 184 (39.32%) men.

CONCLUSIONS

The prevalence rate of CACs detected on panoramic radiographs in patients aged 54 and older was 21.68%, which makes it an important clinical problem.

Arterial Calcification and Its Association With Stroke: Implication of Risk, Prognosis, Treatment Response, and Prevention

Stroke is a leading cause of death worldwide. Vascular calcification (VC), defined as deposition of calcium-phosphate complexes in the vessels, is considered as the characteristic of vascular aging. Calcifications at different vessel layers have different implications. Intimal calcification is closely related to atherosclerosis and affects plaque stability, while medial calcification can cause arterial stiffening and reduce compliance. Accumulating evidence suggested that arterial calcifications, including calcifications in the intracranial artery, coronary artery, and carotid artery, are associated with the risk, prognosis, and treatment response of stroke. VC can not only serve as a marker of atherosclerosis, but cause cerebral hemodynamic impairment. In addition, calcifications in large arteries are associated with cerebral small vessel disease. In this review, we summarize the findings of recently published studies focusing on the relationship between large artery calcification and the risk, prognosis, treatment response, and prevention of stroke and also discuss possible mechanisms behind those associations.

Evaluation of maxillary sinusitis from panoramic radiographs and cone-beam computed tomographic images using a convolutional neural network

Purpose

Materials and Methods

Results

Conclusion

Prevalence and features of elongated styloid process on imaging studies: a systematic review and meta-analysis

OBJECTIVE

To determine the prevalence of the elongated styloid process (ESP) and its characteristics, such as sex and age of the patient, unilateral and bilateral incidence, besides variations between different populations and panoramic and CBCT examinations.

MATERIALS AND METHODS

A search was performed in six databases (PubMed, Web of Science, Scopus, Cochrane, Lilacs, and Embase) to identify observational studies that used imaging exams and assessed ESP prevalence among panoramic radiograph CBCT examinations, whose transversal prevalence studies were included. Furthermore, studies with a specific group of patients or symptomatic patients were excluded. Additionally, Joanna Briggs Institute checklist was used to evaluate the quality of the studies. A meta-analysis was conducted, then subgroup analyses were performed by grouping studies according to the secondary outcomes, with a significance level set at 5%. The Grading of Recommendations Assessment, Development, and Evaluation system was used to rate the certainty in the evidence.

RESULTS

The initial search resulted in 1635 studies, from which 39 articles met the inclusion criteria, encompassing 50,655 participants. The sample size varied between 82 and 5,000 participants. The prevalence of the ESP ranged from 1.3 to 94.8%, with an overall prevalence of 30.2%. The bilateral occurrence was higher than the unilateral one, but no significant predilection was observed according to sex, age, or population. The type of imaging examination also showed no difference in its detection.

CONCLUSION

The overall prevalence of ESP was 30.2%, with a propensity for bilaterality, but not for any sex, age, or population geographic location. The imaging examination modality did not influence the diagnosis of ESP. However, the quality level of the studies evaluated was very low, demonstrating the need for more homogeneous primary studies on the prevalence of the ESP with a more standardized methodology.

CLINICAL RELEVANCE

There is no consensus in the literature regarding the prevalence of the ESP and the characteristics of the affected patients that can cause chronic and debilitating discomfort in the head and neck region. Therefore, knowledge about the prevalence and characteristics of this condition would help dental clinicians reach the correct diagnosis.

Prevalence of soft tissue calcifications in panoramic radiographs of the maxillofacial region of older adults

OBJECTIVES

To investigate the presence of soft tissue calcifications in the head and neck region on panoramic radiographs of older adults.

METHODS

We analysed 1176 panoramic radiographs obtained between January 2013 and December 2018 from individuals of both sexes aged 60 years or older, who were referred by dental specialities to the Dental Imaginology Service of the Federal University of Rio Grande do Norte, Brazil. The types of soft tissue calcification evaluated were as follows: carotid artery calcification (CAC), thyroid cartilage calcifications, triticeous cartilage calcifications, sialoliths, tonsilloliths and lymph node calcifications. The presence of soft tissue calcifications was correlated with age and sex. The chi-square test with continuity correction was used for the calculation of p values and the evaluation of the proposed associations. Prevalence ratios and 95% confidence intervals were also calculated.

RESULTS

At least one type of soft tissue calcification was found in 43% of the sample. The main calcifications detected were CAC, thyroid and triticeous cartilage calcifications, tonsilloliths, sialoliths, calcified lymph nodes, and phleboliths. Mean patient age was 67.47 years and there was a predominance of females (62.8%) in the sample. Bivariate analysis showed a statistically significant association between female sex and the presence of thyroid and triticeous cartilage calcifications and between male sex and the presence of tonsilloliths.

CONCLUSION

Routine panoramic radiography permits the identification of soft tissue calcifications that may be indicators of future cardiovascular disorders, the referral to a medical service and the establishment of therapies for stroke prevention.

Radiological evaluation of maxillofacial soft tissue calcifications with cone beam computed tomography and panoramic radiography

OBJECTIVE

We aimed to present the radiologic characteristics of maxillofacial soft tissue calcifications with a comparison of cone beam computed tomography (CBCT) and panoramic radiography (PR) findings.

MATERIALS AND METHODS

The study was based on CBCT images obtained for different purposes between October 2017 and September 2018. The absence, location and radiological characteristics of some calcifications were evaluated in 252 patients. Statistical analysis was performed with SPSS version 21.0; P < .05 was considered to indicate statistical significance.

RESULTS

Palatine tonsilloliths were the most common calcifications in our study. Calcifications were most frequent in the sixth decade of life. Women had larger calcifications than men. There was a relationship between the sizes measured by CBCT and PR for tonsilloliths. In PR, Region 2 had significantly more calcifications compared with the other regions. The C2 vertebral level was the most common region for tonsilloliths based on CBCT.

CONCLUSIONS

Tonsilloliths have a high prevalence. The regular peripheral type had a higher risk of being a tonsillolith in PR. The peripheral/internal characteristics and the dimensions of calcifications between the two imaging modalities were in harmony. The location of calcification in PR and CBCT was important to distinguish the type of calcification.

Can carotid artery calcifications on dental radiographs predict adverse vascular events? A systematic review

OBJECTIVES

To investigate the predictive value of carotid artery calcifications (CACs) seen on dental imaging for future adverse events related to cerebrovascular and cardiovascular diseases.

METHODS

Electronic databases (PubMed, Embase, SCOPUS and Cochrane) and selected grey literature were searched. Outcomes included stroke, transient ischaemic attack (TIA), myocardial infarction (MI), angina, heart failure, future interventional revascularization procedures and death. Risk of bias was assessed using the Newcastle-Ottawa Scale.

RESULTS

5 studies were selected from 1011 records. All were of moderate to low risk of bias. Results were heterogeneous but showed that patients with CACs on panoramic radiographs (PANs) were more likely to experience strokes, TIA, MI and future revascularization procedures compared to control groups. The differences between groups were not all found to be statistically significant. Bilateral vessel-outlining CACs were an independent risk marker for future vascular events. Patients who are not currently being managed for cardiovascular risk factors should be referred for further evaluation. Those already being treated for atherosclerosis may not benefit from additional investigations.

CONCLUSION

The evidence for the predictive value of CACs on PANs for adverse future vascular events is equivocal but can help to identify at-risk patients who require further evaluation.

CLINICAL RELEVANCE

Dentists must be able to recognize CACs on PANs and make appropriate referrals for patients to be evaluated for cardiovascular risk factors, especially those who have not previously been assessed.

Performance of deep learning object detection technology in the detection and diagnosis of maxillary sinus lesions on panoramic radiographs

OBJECTIVE

The first aim of this study was to determine the performance of a deep learning object detection technique in the detection of maxillary sinuses on panoramic radiographs. The second aim was to clarify the performance in the classification of maxillary sinus lesions compared with healthy maxillary sinuses.

METHODS

The imaging data for healthy maxillary sinuses (587 sinuses, Class 0), inflamed maxillary sinuses (416 sinuses, Class 1), cysts of maxillary sinus regions (171 sinuses, Class 2) were assigned to training, testing 1, and testing 2 data sets. A learning process of 1000 epochs with the training images and labels was performed using DetectNet, and a learning model was created. The testing 1 and testing 2 images were applied to the model, and the detection sensitivities and the false-positive rates per image were calculated. The accuracies, sensitivities and specificities were determined for distinguishing the inflammation group (Class 1) and cyst group (Class 2) with respect to the healthy group (Class 0).

RESULTS

Detection sensitivities of healthy (Class 0) and inflamed (Class 1) maxillary sinuses were 100% for both testing 1 and testing 2 data sets, whereas they were 98 and 89% for cysts of the maxillary sinus regions (Class 2). False-positive rates per image were nearly 0.00. Accuracies, sensitivities and specificities for diagnosis maxillary sinusitis were 90-91%, 88-85%, and 91-96%, respectively; for cysts of the maxillary sinus regions, these values were 97-100%, 80-100%, and 100-100%, respectively.

CONCLUSION

Deep learning could reliably detect the maxillary sinuses and identify maxillary sinusitis and cysts of the maxillary sinus regions.

ADVANCES IN KNOWLEDGE

This study using a deep leaning object detection technique indicated that the detection sensitivities of maxillary sinuses were high and the performance of maxillary sinus lesion identification was ≧80%. In particular, performance of sinusitis identification was ≧90%.

Deep Neural Networks for Dental Implant System Classification

In this study, we used panoramic X-ray images to classify and clarify the accuracy of different dental implant brands via deep convolutional neural networks (CNNs) with transfer-learning strategies. For objective labeling, 8859 implant images of 11 implant systems were used from digital panoramic radiographs obtained from patients who underwent dental implant treatment at Kagawa Prefectural Central Hospital, Japan, between 2005 and 2019. Five deep CNN models (specifically, a basic CNN with three convolutional layers, VGG16 and VGG19 transfer-learning models, and finely tuned VGG16 and VGG19) were evaluated for implant classification. Among the five models, the finely tuned VGG16 model exhibited the highest implant classification performance. The finely tuned VGG19 was second best, followed by the normal transfer-learning VGG16. We confirmed that the finely tuned VGG16 and VGG19 CNNs could accurately classify dental implant systems from 11 types of panoramic X-ray images.

Automatic detection and classification of radiolucent lesions in the mandible on panoramic radiographs using a deep learning object detection technique

OBJECTIVE

The aim of this study was to investigate whether a deep learning object detection technique can automatically detect and classify radiolucent lesions in the mandible on panoramic radiographs.

STUDY DESIGN

Panoramic radiographs of patients with mandibular radiolucent lesions of 10 mm or greater, including ameloblastomas, odontogenic keratocysts, dentigerous cysts, radicular cysts, and simple bone cysts, were included. Lesion labels, including region of interest coordinates, were created in text format. In total, 210 training images and labels were imported into the deep learning GPU training system (DIGITS). A learning model was created using the deep neural network DetectNet. Two testing data sets (testing 1 and 2) were applied to the learning model. Similarities and differences between the prediction and ground-truth images were evaluated using Intersection over Union (IoU). Sensitivity and false-positive rate per image were calculated using an IoU threshold of 0.6. The detection performance for each disease was assessed using multiclass learning.

RESULTS

Sensitivity was 0.88 for both testing 1 and 2. The false-positive rate per image was 0.00 for testing 1 and 0.04 for testing 2. The best combination of detection and classification sensitivity occurred with dentigerous cysts.

CONCLUSIONS

Radiolucent lesions of the mandible can be detected with high sensitivity using deep learning.