Panoramic Radiographs Research Articles

A deep-learning system for diagnosing ectopic eruption

Objectives: To construct a diagnostic model for mixed dentition using a multistage deep-learning network to predict potential ectopic eruption in permanent teeth by integrating dentition segmentation into the process of automatic classification of dental development stages.Methods: A database was established by reviewing 1576 anonymous panoramic radiographs of children aged 6–12 years, collected at the Stomatology Hospital, Zhejiang University School of Medicine. These radiographs were categorised as normal or ectopic eruption, with expert diagnoses serving as a benchmark for training and evaluating artificial intelligence (AI) models. Furthermore, tooth boundaries and dental development stages were manually annotated by three pediatric dentistry experts. The dataset was split into training, validation, and test sets at an 8:1:1 ratio.Results: The diagnostic performance of the deep-learning model was rigorously evaluated. The model demonstrated accuracy in tooth segmentation, with Intersection over Union, precision, sensitivity, and F1 scores of 0.959, 0.993, 0.966, and 0.979, respectively. Furthermore, its ability to identify tooth ectopic eruptions on panoramic radiographs, when compared to evaluations by three dentists. Based on McNemar's test, the model's specificity and accuracy in identifying ectopic tooth eruptions on the test dataset surpassed that of Dentist 1 (P < 0.05), while no significant difference was observed compared to the other two dentists. Besides, the deep learning model also showed its potential in classifying dental development stages, as tested against three different standards.Conclusions: The adaptability of the AI-enabled model in this study was demonstrated across multiple scenarios, with clinical validation highlighting its efficacy in diagnosing ectopic eruptions using a multistage deep-learning approach.Clinical Significance: Our findings provide new insights and technical support for the prevention and treatment of abnormal tooth eruption, laying the groundwork for predictive models for other prevalent pediatric dentistry conditions.

Radiographic Study of the Prevalence and Pattern of Impacted Mandibular Third Molars in a Northern Nigerian Population.

Background: The incidence of impacted mandibular and maxillary third molars has become a global public health concern. This study reports a radiographic prevalence and pattern of impacted mandibular third molars among a Nigerian population. Objectives: The aim of this study was to determine the prevalence and pattern of impacted mandibular third molars in a Northern Nigeria population. Materials and Methods: This retrospective cross-sectional study was conducted at the Oral Diagnostic Sciences Department of the Aminu Kano Teaching Hospital Kano, Kano State, Nigeria. Digital diagnostic images of patients of record on Planmeca Promax® within the sampling frame and acquired during the two years under review, were included in the study. The images were analyzed on Planmeca Romexis 4.3.0 R software to identify relevant study variables. Descriptive statistics was done using SPSS for windows software version 23.0 (IBM, Chicago IL., USA). Also, Pearson’s Chi‑square (χ2) statistical test was applied while confidence interval and P-value were set at 95% and ≤0.05, respectively. Results: A total of 4,932 pantomographs were reviewed and 576 were selected for the study. There were 824 impacted mandibular third molars within the age range of 18-65 years (mean age±SD = 32.67±9.69). 297 (51.6%) were males, and 279 (48.4%) were females. The prevalence of impacted mandibular third molar was 16.71%. Impacted mandibular third molar occurred more frequently in the 26-35 years age range. The most frequent angle of impaction was horizontal, followed by mesioangular, and the least frequent angulation was distoangular. Conclusion: The prevalence and pattern of impacted third molars among Northern Nigeria population are almost similar to other racial populations with minor variations, and the prevalence decreases with increasing age. A proper radiographic evaluation of the patterns of third molars impaction is, therefore, essential to assist dental surgeons in making decisions with regard to surgical planning and treatment. Key Words: Impacted teeth, panoramic radiograph, third molar.

AI-Powered Identification of Osteoporosis in Dental Panoramic Radiographs: Addressing Methodological Flaws in Current Research.

Background: Osteoporosis, a systemic skeletal disorder, is expected to affect 60% of women over 50. While dual-energy X-ray absorptiometry (DXA) scans are the current gold standard for diagnosis, they are typically used only after fractures occur, highlighting the need for early detection tools. Initial studies have shown panoramic radiographs (PRs) to be a potential medium, but these have methodological flaws. This study aims to address these shortcomings by developing a robust AI application for accurate osteoporosis identification in PRs. Methods: A total of 348 PRs were used for development, 58 PRs for validation, and 51 PRs for hold-out testing. Initially, the YOLOv8 object detection model was employed to predict the regions of interest. Subsequently, the predicted regions of interest were extracted from the PRs and processed by the EfficientNet classification model. Results: The model for osteoporosis detection on a PR achieved an overall sensitivity of 0.83 and an F1-score of 0.53. The area under the curve (AUC) was 0.76. The lowest detection sensitivity was for the cropped angulus region (0.66), while the highest sensitivity was for the cropped mental foramen region (0.80). Conclusion: This research presents a proof-of-concept algorithm showing the potential of deep learning to identify osteoporosis in dental radiographs. Furthermore, our thorough evaluation of existing algorithms revealed that many optimistic outcomes lack credibility when subjected to rigorous methodological scrutiny.

Deep learning for diagnostic charting on pediatric panoramic radiographs.

Artificial intelligence (AI)-based systems are used in dentistry to ensure a more accurate and efficient diagnostic process. The objective of the present study was to evaluate the performance of a deep learning (DL) program for the detection and classification of dental structures and treatments on panoramic radiographs of pediatric patients. In total, 4821 anonymized digital panoramic radiographs of children between 5 and 13 years of age were analyzed by YOLOv4, a CNN (Convolutional Neural Networks)-based object detection model. The ability to make a correct diagnosis was tested on samples from pediatric patients examined within the scope of the study. All statistical analyses were performed using SPSS version 26.0 software. The YOLOv4 model diagnosed the primary teeth, permanent tooth germs, and brackets successfully, with high F1 scores of 0.95, 0.90, and 0.76, respectively. Although this model achieved promising results, there were certain limitations for some dental structures and treatments, including fillings, root canal treatments, and supernumerary teeth. The architecture of the present study achieved reliable results, with some specific limitations for detecting dental structures and treatments. The detection of certain dental structures and previous dental treatments on pediatric panoramic radiographs by using a DL-based approach may provide early diagnosis of some dental anomalies and help dental practitioners to find more accurate treatment options by saving time and effort.

Osteoporotic alterations of mandible in aged women using panoramic radiograph: A multicentric study

Osteoporotic alterations of mandible in aged women using panoramic radiograph: A multicentric study

Simple bone cyst of the mandible: A case report with review of literature

Simple bone cyst (SBC) was first described by Lucas in 1929. It is an asymptomatic, intraosseous cyst. It falls under pseudocyst category, as it is devoid of an epithelial lining and mostly discovered accidently on routine radiograph. It commonly affects metaphyses of long bones such as humerus and femur. Other terminologies used for this lesion include Solitary bone cyst, traumatic bone cyst and hemorrhagic bone cyst. The incidence of cyst affecting the jaws is only 1%. About 89% of the lesion occurs in mandible and 11% in maxilla. The purpose of this article is to present rare occurrence of SBC in the mandibular region with review of literature. An 18 year old male patient reported with pain in right lower back teeth region. Clinically 47 was tender. Panoramic radiograph showed a radiolucent unilocular lesion in ascending ramus and mandibular body distal to 47. Diagnosis of apical periodontitis in relation to 47 and simple bone cyst mandible was made by correlating clinical, radiographic and histopathological evaluation. Surgical curettage was performed and follow up after 4 months revealed no evidence of recurrence. Simple bone cyst is rare incidental lesion of the mandible. Thorough clinical, radiological and histopathologic evaluation is helpful for diagnosis and surgical management of the lesion. Conservative management should be approached to prevent noble structure, with follow up to confirm remission of the lesion.

Comparative evaulation of condylar guidance obtained by digital panoramic radiographic images with that of obtained during jaw relation and try-in using two different interocclusal record material in completely edentulous patients: an in vivo study

Background In completely edentulous patients, maintaining the patient’s temporomandibular joint (TMJ) health and intraoral components is very essential, which can be achieved by an accurate interocclusal record, which can be taken by various materials and also at various stages. Simulation of the temporomandibular joint anatomy and mandibular movements is possible on a semi-adjustable articulator, which is widely used in prosthodontics. Transferring the record on articulator is a critical step in the fabrication of complete dentures, so it should be done precisely to reduce chair side time and also to avoid complications for patients. This study examines which material is more accurate and records which stage gives more precision by comparing the Condylar guidance on radiographs. This is because it is a relatively consistent bony landmark and can be used to correlate with clinical methods in terms of verification. The objective is to determine the inter occlusal records created during the Jaw Relation stage and Try-in using reinforced wax and Polyvinyl siloxane and compare these values to the condylar guidance obtained by orthopantomogram (OPG). Methods Recording the Condylar Guidance value on OPG by tracing the Glenoid fossa, Articular eminence, and the Frankfurt Horizontal Plane. Followed by conventional steps in making complete dentures at jaw relation and try-in stage. We will record values of condylar guidance with the two different Interocclusal record materials. Expected result It will help the clinicians to gain knowledge about obtaining precise condylar inclination angle by using suitable interocclusal record materials. Conclusions This study will help us to obtain condylar inclination, which in turn helps the patient’s satisfaction as it preserves the intraoral components and health of TMJ.

Frank’s Sign which is a Marker of Coronary Heart Disease

In today's clinical practice, the importance of a thorough clinical examination of the patient is emphasised in the article. There are 2 real-world instances of diagonal earlobe creases, commonly referred to as Frank's sign. The patients in both cases have ischemic heart disease, however they differ in terms of age and Frank's sign severity. Data from the literature are presented regarding the frequency of Frank's disease in various population groups, as well as its clinical importance and potential underlying pathophysiological causes. In the medical literature, the diagonal ear lobe crease (DELC) is termed as a substitute sign that can detect high-risk individuals with hidden atherosclerosis. However, there hasn't been any research on this subject in the dentistry or medical literature. Most clinical, angiography, and postmortem data are in favour of the idea that DELC is a useful extravascular physical marker that can determine individuals who are at risk of dying from coronary artery atherosclerosis. Some research findings have failed to support the theory linking DELC to certain health conditions. However, recent studies utilizing B mode ultrasonography have indicated a connection between DELC and the development of carotid artery atherosclerosis. Another study has also associated DELC with the existence of calcified carotid artery atheromas visible on panoramic radiographs. Notably, DELC is highly noticeable on head and neck oncology screening examinations. When the patient's vital signs, medical examination, and panoramic radiograph are combined, the results are DELC can aid to assess the patient's risk of atherosclerosis.

Evaluation of the accuracy of digital indirect bonding vs. conventional systems: a randomized clinical trial.

To compare the accuracy and chair time of self-ligating brackets using direct bonding, traditional indirect bonding (IB), and computer-aided design/computer-aided manufacturing (CAD/CAM) IB techniques after orthodontic leveling and alignment. Forty-five patients were randomly assigned to three bonding groups (G1 [n = 15], G2 [n = 15], and G3 [n = 15]). Evaluation after the alignment and leveling phases used two parameters of the objective grading system of the American Board of Orthodontics for root parallelism and posterior marginal ridges, assessed using panoramic radiographies (PR I and PR II), a digital model, and a plaster model. Blinding was only applied for outcome assessment. No serious harm was observed except for gingivitis associated with plaque accumulation. Although G3 showed better numerical results, they were not statistically significant in the radiographic or model evaluations (P > .001). Mean chair time was significantly shorter in G3 (1.1 ± 11.8 min) vs. G1 (56.7 ± 7.3 min) and G2 (52.8 ± 8.3 min; P < .001). The CAD/CAM IB system for self-ligating brackets was as effective as conventional methods, with a shorter chair time.

Oro-Dental Characteristics in Patients With Adult-Onset Hypophosphatasia Compared to a Healthy Control Group-A Case-Control Study.

Hypophosphatasia (HPP) is a rare inherited disease that affects multiple organ systems including bone and teeth. Limited knowledge exists on dental and oral health in patients with adult-onset HPP (aHPP). The aim of this study was to investigate oro-dental characteristics in patients with aHPP compared to healthy controls. This case-control study included 20 patients with aHPP compared to 31 healthy controls. Oro-dental manifestations were examined by standardised interviews, clinical examinations as well as radiological registrations on panoramic radiograph (OP) and cone-beam computed tomography (CBCT) scans. The subjective experience of tooth fractures (p = 0.010), caries in permanent teeth (p = 0.032) and early loss of permanent teeth (p = 0.002) was significantly higher in patients with aHPP compared to the controls. In the aHPP group, the presence of specific teeth (p ≤ 0.045) and attrition of 11 were significantly lower (p = 0.012) compared to the controls. Opacity of a few teeth (p ≤ 0.049), presence of denticles (p = 0.024), the distance between the enamel-cement junction (CEJ) and the marginal bone level at specific sites (p ≤ 0.021) and crown height of 11 (p = 0.017) were significantly higher in patients with aHPP than in healthy controls. The results indicate that patients with aHPP have a subjective experience of having poorer dental health. Loss of permanent teeth, less attrition, tooth opacities, denticles and larger distance between CEJ and marginal bone level are possible oro-dental findings in patients with aHPP.

Unleashing the potential of applied UNet architectures and transfer learning in teeth segmentation on panoramic radiographs

Accurate tooth segmentation in panoramic radiographs is a useful tool for dentists to diagnose and treat dental diseases. Segmenting and labeling individual teeth in panoramic radiographs helps dentists monitor the formation of caries, detect bone loss due to periodontal disease, and determine the location and orientation of damaged teeth. It can also aid in both the planning and placement of dental implants, as well as in forensic dentistry for the identification of individuals in criminal cases or human remains. With the advancement of artificial intelligence, many deep learning-based methods are being developed and improved. Although convolutional neural networks have been extensively used in medical image segmentation, the UNet and its advanced architectures stand out for their superior segmentation capacities. This study presents four semantic segmentation UNets (Classic UNet, Attention UNet, UNet3+, and Transformer UNet) for accurate tooth segmentation in panoramic radiographs using the new Tufts Dental dataset. Each model was performed using transfer learning from ImageNet-trained VGG19 and ResNet50 models. The models achieved the best results compared to the other literature models with dice coefficients (DC) and intersection over union (IoU) of 94.64% to 96.98% and 84.27% to 94.19%, respectively. This result suggests that Unet and its variants are more suitable for segmenting panoramic radiographs and could be useful for potential dental clinical applications.

A dual-labeled dataset and fusion model for automatic teeth segmentation, numbering, and state assessment on panoramic radiographs

BackgroundRecently, deep learning has been increasingly applied in the field of dentistry. The aim of this study is to develop a model for the automatic segmentation, numbering, and state assessment of teeth on panoramic radiographs.MethodsWe created a dual-labeled dataset on panoramic radiographs for training, incorporating both numbering and state labels. We then developed a fusion model that combines a YOLOv9-e instance segmentation model with an EfficientNetv2-l classification model. The instance segmentation model is used for tooth segmentation and numbering, whereas the classification model is used for state evaluation. The final prediction results integrate tooth position, numbering, and state information. The model’s output includes result visualization and automatic report generation.ResultsPrecision, Recall, mAP50 (mean Average Precision), and mAP50-95 for the tooth instance segmentation task are 0.989, 0.955, 0.975, and 0.840, respectively. Precision, Recall, Specificity, and F1 Score for the tooth classification task are 0.943, 0.933, 0.985, and 0.936, respectively.ConclusionsThis fusion model is the first to integrate automatic dental segmentation, numbering, and state assessment. It provides highly accurate results, including detailed visualizations and automated report generation.

Prevalence and predictive parameters of external root resorption caused by retained wisdom teeth

ObjectivesThe aim of this study was to estimate the prevalence and severity of external root resorption (RR) caused by retained third molars (M3), to compare the sensitivity of panoramic radiography (PAN) and cone beam computed tomography (CBCT) and to determine predictive factors for root resorption.Materials and methodsIn a retrospective cross-sectional analysis, we included patients (N = 367) who underwent PAN and CBCT imaging between December 2017 and July 2019. Previous orthodontic treatment, age, gender, superimposition of second molars (M2) and M3 on PAN, retention depth, inclination angle and vertical level of contact with the M2 were used as predictor variables. The outcome variable was RR of the M2, graded according to Ericson et al. [1]. Subgroup analyses compared patients with and without suspected resorption in the PAN.ResultsWhile less than 5% of PANs suggested RR associated with M3, CBCT showed RR in 20% of all M2 with adjacent retained M3. The angle of inclination of M3, patient age and vertical level of molar contact emerged as predictive parameters, with mesial inclination, older age and deeper retention associated with increased severity of M2.ConclusionWithin the limitations of our study, these data confirm the poor performance of PAN in the diagnosis of RR. CBCT may be helpful in detecting RR in mesioangulated and deeply retained M3 in elderly patients, even when PAN did not suggest pathology.Clinical relevanceOur study may help to decide whether CBCT should be considered prior to M3 surgery.

Evaluation of fractal analysis and radiomorphometric measurements of mandibular bone structure in bruxism and non-bruxism paediatric patients.

The goal of this examination was to compare the impact of probable sleep/awake bruxism on the mandibular trabecular bone structure by fractal analysis (FA) with digital panoramic radiograph (DPR) and radiomorphometric measurements in paediatric patients with bruxism. The examination included 130 participants with 63 patients with probable sleep/awake bruxism and 67 control groups. Bilateral regions of interest (ROI) in three regions were examined as ROI1: mandibular ramus, ROI2: mandibular angulus, ROI3: anterior to the molar teeth. Radiomorphometric measurements were taken of the mandibular cortical width (MCW), panoramic mandibular index (PMI), and mandibular cortical index (MCI). p < 0.05 was approved for statistical significance. The ROI-1, ROI-2, and ROI-3 values were defined to be statistically significantly high in the bruxism group (p < 0.05). No significant difference was found between the groups in the other values (p > 0.05). There was no difference in the age and gender for any of the parameters (p > 0.05). In children and adolescents, the mandibular trabecular bone can be affected by bruxism. FA can be used as an auxiliary method for finding the mandibular trabecular differences of patients with bruxism in paediatric dentistry just as it can for adults.

Does the Timing of Orthognathic Surgery Affect the Level of External Root Resorption?: Surgery-First Approach Versus Conventional Orthognathic Surgery.

This study aimed to compare external root resorption levels, both linearly and areally, on panoramic radiographs in Class III patients whose treatment was completed with the conventional orthognathic surgery and the surgery-first approach. The study included 20 patients (10 female, 10 male; 23.98±2.63y) treated with conventional orthognathic surgery and 18 patients (8 female, 10 male; 27.5±3.42y) treated with surgery-first approach was included. Linear and area measurements were made on the digital panoramic radiographs taken before and at the end of the treatment of 38 patients who constituted the sample of the study. The mean changes observed in each group were evaluated using a paired t test. The chronological ages, skeletal features (type of Class III malocclusion), treatment durations, initial values, and mean changes that occurred in each group were compared using independent t test. When the changes in root lengths and root surface areas were evaluated in all teeth in both groups, a significant loss in size was measured (P<0.001). More resorption was observed in teeth 12 and 22 in the conventional surgery group (P<0.05). When the amount of resorption in both groups was compared, there was significantly more resorption in tooth number 12 in the conventional surgery group (P<0.05). The study showed that there was external root resorption in all teeth in both protocols, but the root lengths of the lateral teeth were shortened more in both groups.

Impacted third molar surgery in older patients—Is patient´s age really a risk factor for complications?

ObjectivesThe aim of this study was to analyze the influence of patients´ age on perioperative complications in impacted third molar surgery and how established risk factors are affected by age.Materials and methodsThe clinical findings, digital panoramic radiographs and perioperative data of 200 patients (554 impacted third molars) that had been subjected to tooth extraction, from July 2023 until July 2024, were analyzed. Perioperative complications (Inferior alveolar nerve (IAN) hypesthesia, oroantral communication (OAC), lingual nerve (LN) hypesthesia, postoperative bleeding, postoperative infection) as well as impaction patterns and risk factors (angulation type, bone coverage, depth- and risk scores) were analyzed by age (cut-off 30 years).ResultsThe population was divided into two groups by age (Group A = ≥ 30 years (n = 52) vs. Group B = < 30 years (n = 148)). Upper third molars showed significantly deeper bone coverage, higher depth scores, higher risk scores and different angulation types in patients aged < 30 years. Mandibular third molars showed significantly deeper bone coverage, higher depth scores, higher risk scores according and different angulation types in patients aged ≥ 30 years. However, IAN hypesthesia, LN hypesthesia, postoperative bleeding and postoperative infection did not show any significant differences regarding patients’ age.ConclusionThe current findings suggest that age (cut-off 30 years) does not statistically correlate with a higher risk for postoperative complications in impacted third molar surgery in contrast to recent publications.Clinical relevanceIn contrast to recent publications, the present study falsified a positive correlation between patients’ age and the occurrence of postoperative complications in impacted third molar surgery. Therefore, other risk factors should be investigated in order to minimize these procedure specific complications.

7508 Primary Hyperparathyroidism Caused By A CDC73-related Atypical Parathyroid Tumor

Abstract Disclosure: R. Cardenas: None. S. Tariq: None. R. Habibi: None. N. Ebrahimi: None. E. Astiazaran-Symonds: None. Background: Primary hyperparathyroidism (PHPT) is a common endocrine disease, causes include: parathyroid adenomas (80-85%), parathyroid hyperplasia (10-15%), atypical parathyroid adenomas (APA) (1.2 %) and parathyroid carcinomas (PC) (1%). APA share many anatomic and histopathologic features with PC making the distinction between the two challenging. Although most are sporadic, some APA are caused by germline mutations in the CDC73 gene, which is associated with hyperparathyroidism-jaw tumor (HPT-JT) syndrome PC and familial isolated PHPT with APA. Clinical Case: A 42-year-old female presented to the clinic for evaluation of recurrent PHPT. She was diagnosed at age 22 when she experienced fatigue, polydipsia with hypercalcemia (12.9mg/dl). She underwent superior bilateral parathyroidectomy with pathology consistent with parathyroid adenoma. Twenty years later, symptoms recurred and tests showed hypercalcemia (13.0 mg/dl), elevated PTH (252 pg/dl), kidney function preserved (GFR: 82.5 ml/min), normal vitamin D 25 OH (31 ng/dl). Patient denied history of nephrolithiasis or fractures. Family history was positive for kidney stones in father and siblings, also history of a jaw tumor in niece. Sestamibi scan showed a left inferior parathyroid adenoma. Patient underwent resection of the adenoma; pathology revealed a left parathyroid adenoma with atypical parathyroid tumor, &amp;gt;95% cellularity with cytologic features concerning for malignancy, prominent macronucleoli, necrosis, elevated mitotic activity and Ki67 with 15% tumor cells without vascular invasion. Post-op Calcium and PTH normalized (9.6 mg/dl and 10.8 pg/ml, respectively). Genetic testing via multi-gene panel identified a whole gene deletion of CDC73, consistent with hyperparathyroidism-jaw tumor syndrome. Panoramic jaw x-rays did not show jaw tumors. Genetic testing for family members is underway. Conclusion: Although somatic mutations in CDC73 are frequently found in sporadic APA and PC, germline mutations are a rare cause of PHPT associated with these tumors. Histopathological examination is necessary to confirm the diagnosis of APA and to differentiate it from PC, which has a recurrence rate &amp;gt;50%. APA should be treated as a “tumor of uncertain malignant potential” and needs to be carefully and periodically followed up with measurement of Ca, PTH and vitamin D. Confirmation of a CDC73 mutation by genetic testing helped identify the increased risk for other associated tumors in jaw, kidneys and uterus prompting initiation of surveillance with panoramic jaw x-rays with neck shielding, renal US and pelvis US. Genetic testing of family members will allow timely detection and treatment of individuals at risk. Presentation: 6/3/2024

Do Tunisian children with unilateral cleft lip and palate differ from normal individuals in dental maturity? A pilot study.

Cleft lip and palate (CLP) is the most common congenital malformation of the head and neck. Children with CLP often exhibit dental anomalies. To evaluate the dental age (DA) of unilateral CLP in Tunisian children. This was a cross-sectional study carried out in the department of pediatric dentistry at the University Hospital La Rabta, Tunis. Patients aged between 5 and 14 years, with no other congenital anomalies or syndromes in the craniofacial region other than CLP, were included. The patients' chronological ages were first calculated in years and months. DA was assessed in panoramic radiographs using Demirjian's method. The score of each stage is allocated, and the sum of the scores provides an evaluation of the subject's dental maturity. Fifty-three patients were included in the present study. No difference was observed between the two groups regarding the dental age. A strong and positive correlation between the DA and the chronological age in the two groups was observed (r=0.826). Estimated regression showed that chronological age alone explained 57,4% (r2=0.574) of the dental age variation in the study group and 64.5% (r2=0.645) in the control group. For dental management, CLP children should have the same approach in orthodontics and pediatric dentistry as individuals without clefts, with a focus on the individualization of diagnosis and treatment planning.

Unsupervised few shot learning architecture for diagnosis of periodontal disease in dental panoramic radiographs.

In the domain of medical imaging, the advent of deep learning has marked a significant progression, particularly in the nuanced area of periodontal disease diagnosis. This study specifically targets the prevalent issue of scarce labeled data in medical imaging. We introduce a novel unsupervised few-shot learning algorithm, meticulously crafted for classifying periodontal diseases using a limited collection of dental panoramic radiographs. Our method leverages UNet architecture for generating regions of interest (RoI) from radiographs, which are then processed through a Convolutional Variational Autoencoder (CVAE). This approach is pivotal in extracting critical latent features, subsequently clustered using an advanced algorithm. This clustering is key in our methodology, enabling the assignment of labels to images indicative of periodontal diseases, thus circumventing the challenges posed by limited datasets. Our validation process, involving a comparative analysis with traditional supervised learning and standard autoencoder-based clustering, demonstrates a marked improvement in both diagnostic accuracy and efficiency. For three real-world validation datasets, our UNet-CVAE architecture achieved up to average 14% higher accuracy compared to state-of-the-art supervised models including the vision transformer model when trained with 100 labeled images. This study not only highlights the capability of unsupervised learning in overcoming data limitations but also sets a new benchmark for diagnostic methodologies in medical AI, potentially transforming practices in data-constrained scenarios.

Fractal Analysis of Mandible in Panoramic Radiographs of Patients Received Radiotherapy for Nasopharyngeal Carcinoma

Purpose: This study aimed to assess the impact of radiotherapy on the internal structure complexity of mandibular cortical and trabecular bone and to determine the duration required for a return to healthy values post-radiotherapy.Materials and Methods: Panoramic radiographs from patients undergoing radiotherapy for nasopharyngeal carcinoma were analyzed before and after treatment. Four groups were formed based on post-radiotherapy radiography timing (0-6 months, 6-12 months, 12-24 months, and 24-36 months), comprising a total of 59 cases and 118 radiographs. Fractal analysis was conducted on four bilateral regions (ROI) in both trabecular and cortical bone on each radiograph. Additionally, measurements of inferior alveolar canal width and mandibular cortical width were performed. Mean and maximum radiation dose values to the mandible were measured, and their correlation with changes in fractal dimension, inferior alveolar canal width, and mandibular cortical width values was assessed.Results: Fractal dimension values in regions over trabecular bone showed a statistically significant decrease in all groups, although no significant difference was observed among the four groups. In ROI-4 from cortical bone, a significant fractal dimension decrease was noted in all groups except the 0-6 month group. The magnitude of fractal dimension decrease was higher in the 12-24 and 24-36 month groups compared to the 0-6 month group. inferior alveolar canal width and mandibular cortical width values significantly decreased post-radiotherapy in all groups, with a consistent decrease across the groups.Conclusions: Radiotherapy induces a reduction in the internal complexity of trabecular and cortical bone structures in the mandible. Osteoradionecrosis risk persists even three years post-radiotherapy, suggesting a cautious approach to interventional procedures on the bone.