A YOLO-V5 approach for the evaluation of normal fillings and overhanging fillings: an artificial intelligence study.

Dental fillings, frequently used in dentistry to address various dental tissue issues, may pose problems when not aligned with the anatomical contours and physiology of dental and periodontal tissues. Our study aims to detect the prevalence and distribution of normal and overhanging filling restorations using a deep CNN architecture trained through supervised learning, on panoramic radiography images. A total of 10480 fillings and 2491 overhanging fillings were labeled using CranioCatch software from 2473 and 1850 images, respectively. After the data obtaining phase, validation (80%), training 10%), and test-groups (10%) were formed from images for both labelling. The YOLOv5x architecture was used to develop the AI model. The model's performance was assessed through a confusion matrix and sensitivity, precision, and F1 score values of the model were calculated. For filling, sensitivity is 0.95, precision is 0.97, and F1 score is 0.96; for overhanging were determined to be 0.86, 0.89, and 0.87, respectively. The results demonstrate the capacity of the YOLOv5 algorithm to segment dental radiographs efficiently and accurately and demonstrate proficiency in detecting and distinguishing between normal and overhanging filling restorations.

Evaluation of tooth development stages with deep learning-based artificial intelligence algorithm.

BACKGROUND: This study aims to evaluate the performance of a deep learning system for the evaluation of tooth development stages on images obtained from panoramic radiographs from child patients. METHODS: The study collected a total of 1500 images obtained from panoramic radiographs from child patients between the ages of 5 and 14 years. YOLOv5, a convolutional neural network (CNN)-based object detection model, was used to automatically detect the calcification states of teeth. Images obtained from panoramic radiographs from child patients were trained and tested in the YOLOv5 algorithm. True-positive (TP), false-positive (FP), and false-negative (FN) ratios were calculated. A confusion matrix was used to evaluate the performance of the model. RESULTS: Among the 146 test group images with 1022 labels, there were 828 TPs, 308 FPs, and 1 FN. The sensitivity, precision, and F1-score values of the detection model of the tooth stage development model were 0.99, 0.72, and 0.84, respectively. CONCLUSIONS: In conclusion, utilizing a deep learning-based approach for the detection of dental development on pediatric panoramic radiographs may facilitate a precise evaluation of the chronological correlation between tooth development stages and age. This can help clinicians make treatment decisions and aid dentists in finding more accurate treatment options.

Experiences and Health Outcomes of Emerging Adults with Type 1 Diabetes: A Mixed Methods Study.

BACKGROUND: Emerging adults with type 1 diabetes are at risk of poorer diabetes-related health outcomes than other age groups. Several factors affecting the health and experiences of the emerging adults are culture and healthcare specific. OBJECTIVES: The aim of this study was to explore the experience of emerging adults living with type 1 diabetes in Lebanon, describe their diabetes self-care and diabetes-related health outcomes (HbA1c and diabetes distress), and identify the predictors of these outcomes. METHODS: A convergent mixed methods design was used with 90 participants aged 18-29 years. Sociodemographic, clinical data, and measures of diabetes distress, social support, and self-care were collected. Fifteen emerging adults participated in individual semi-structured interviews. Multiple linear regression was used to determine predictors of diabetes outcomes. Thematic analysis was used to analyze qualitative data. Data integration was used to present the mixed methods findings. RESULTS: The study sample had a mean HbA1c of 7.7% (SD = 1.36) and 81.1 % reported moderate to severe diabetes distress levels. The participants had good levels of diabetes self-care and high levels of social support. HbA1c was predicted by insulin treatment type, age at diagnosis, and diabetes self-care; while diabetes distress was predicted by diabetes knowledge, blood glucose monitoring approach, and diabetes self-care. "Living with type 1 diabetes during emerging adulthood: the complex balance of a chemical reaction" was the overarching theme of the qualitative data, with three underlying themes: "Breaking of bonds: changes and taking ownership of their diabetes", "The reactants: factors affecting the diabetes experience", and "Aiming for equilibrium". The integrated mixed methods results revealed one divergence between the qualitative and quantitative findings related to the complexity of the effect of received social support. DISCUSSION: The suboptimal health of the emerging adults despite good self-care highlights the importance of addressing cultural and healthcare specific factors such as diabetes knowledge and public awareness, social support, and availability of technology to improve diabetes health. Findings of this study can guide future research, practice, and policy development.

The role of ceranib-2 and its nanoform on the decrease of telomerase levels in human non-small cell cancer.

BACKGROUND: Ceranib-2, an acid ceramidase (AC) inhibitor, can inhibit cancer cell proliferation and tumor development. However, poor water solubility and low cellular bioavailability limit its efficacy in cancer treatment. METHODS AND RESULTS: This study aimed to investigate the cell death induced by ceranib-2 and its solid lipid nanoformulation (ceranib-2-SLN) produced by the hot homogenization technique and the synergistic relationship between ceramide and telomerase in vitro and in silico. Furthermore, this study proved the possible mechanism of ceranib-2-induced AC inhibition by in silico studies. The effective cytotoxic concentrations of ceranib-2, telomerase level, and changes in ceramide levels were measured by MTT colorimetric cytotoxicity assay, ELISA, and LC/MS/MS methods, respectively. TEM results showed that ceranib-2-SLN was 13-fold smaller than the size of ceranib-2. Ceranib-2 and ceranib-2-SLN had IC50 concentrations of 31.62 (± 2.1) and 27.69 (± 1.75) µM in A549, and 48.79 (± 1.56) and 67.98 (± 2.33) in Beas-2B cells. These compounds simultaneously increased ceramide levels and decreased telomerase levels in A549 cells. Ceranib-2 increased telomerase levels while decreasing ceramide levels in Beas-2B cells. It was shown how the synergistic impact of ceranib-2-induced ceramide production and ceramide-induced telomerase level reduction on cytotoxicity in A549 cells. CONCLUSIONS: Ceranib-2-SLN was discovered to be more cytotoxic on cancer cells than ceranib-2, suggesting that it could be a promising option for the development of a new anti-cancer agent.

Detection of the Fixed Prostheses on Panoramic Images: An Artificial Intelligence Based Study.

OBJECTIVE: To investigate the effectiveness of using YOLO-v5x in detecting fixed prosthetic restoration in panoramic radiographs. STUDY DESIGN: Descriptive study. Place and Duration of the Study: Department of Oral and Maxillofacial Radiology, Eskisehir Osmangazi University, Eskisehir, Turkiye from November 2022 to April 2023. METHODOLOGY: For the labelling of fixed prosthetic restorations, 8,000 panoramic radiographs were evaluated using the YOLO-v5x architecture. In creating the dataset for this study, fixed prosthetic restorations were categorised as dental implant, pontic, crown, and implant-supported crown on dental panoramic radiographs. The labelled images were then randomly split into three groups: 80% for training, 10% for validation, and 10% for testing. The labelled panoramic images constituted the model's training dataset, and leveraging the knowledge acquired during this learning stage, the model generated predictions in the testing phase. RESULTS: The majority of labelling data were dedicated to crown restorations. The precision and sensitivity values of YOLOv5x were 0.99 and 0.98 for crowns, 0.98 and 0.99 for implants, 0.99 and 0.99 for pontics, and 0.99 and 0.99 for implant-supported crowns, respectively. CONCLUSION: The results obtained in this study demonstrate a satisfactory success rate of YOLO-v5x in detecting dental prosthetic restorations. The high precision and sensitivity of the model indicate its strong potential to enhance clinical professional performance and contribute to the development of more efficient dental health services. KEY WORDS: Artificial intelligence, Dentistry, Dental prosthesis, Panoramic radiography.

Automatic deep learning detection of overhanging restorations in bitewing radiographs.

OBJECTIVES: This study aimed to assess the effectiveness of deep convolutional neural network (CNN) algorithms for the detecting and segmentation of overhanging dental restorations in bitewing radiographs. METHODS: A total of 1160 anonymized bitewing radiographs were used to progress the artificial intelligence (AI) system for the detection and segmentation of overhanging restorations. The data were then divided into three groups: 80% for training (930 images, 2399 labels), 10% for validation (115 images, 273 labels), and 10% for testing (115 images, 306 labels). A CNN model known as You Only Look Once (YOLOv5) was trained to detect overhanging restorations in bitewing radiographs. After utilizing the remaining 115 radiographs to evaluate the efficacy of the proposed CNN model, the accuracy, sensitivity, precision, F1 score, and area under the receiver operating characteristic curve (AUC) were computed. RESULTS: The model demonstrated a precision of 90.9%, a sensitivity of 85.3%, and an F1 score of 88.0%. Furthermore, the model achieved an AUC of 0.859 on the receiver operating characteristic (ROC) curve. The mean average precision (mAP) at an intersection over a union (IoU) threshold of 0.5 was notably high at 0.87. CONCLUSIONS: The findings suggest that deep CNN algorithms are highly effective in the detection and diagnosis of overhanging dental restorations in bitewing radiographs. The high levels of precision, sensitivity, and F1 score, along with the significant AUC and mAP values, underscore the potential of these advanced deep learning techniques in revolutionizing dental diagnostic procedures.

A deep learning approach to detection of oral cancer lesions from intra oral patient images: A preliminary retrospective study.

INTRODUCTION: Oral squamous cell carcinomas (OSCC) seen in the oral cavity are a category of diseases for which dentists may diagnose and even cure. This study evaluated the performance of diagnostic computer software developed to detect oral cancer lesions in intra-oral retrospective patient images. MATERIALS AND METHODS: Oral cancer lesions were labeled with CranioCatch labeling program (CranioCatch, Eskisehir, Turkey) and polygonal type labeling method on a total of 65 anonymous retrospective intraoral patient images of oral mucosa that were diagnosed with oral cancer histopathologically by incisional biopsy from individuals in our clinic. All images have been rechecked and verified by experienced experts. This data set was divided into training (n = 53), validation (n = 6) and test (n = 6) sets. Artificial intelligence model was developed using YOLOv5 architecture, which is a deep learning approach. Model success was evaluated with confusion matrix. RESULTS: When the success rate in estimating the images reserved for the test not used in education was evaluated, the F1, sensitivity and precision results of the artificial intelligence model obtained using the YOLOv5 architecture were found to be 0.667, 0.667 and 0.667, respectively. CONCLUSIONS: Our study reveals that OCSCC lesions carry discriminative visual appearances, which can be identified by deep learning algorithm. Artificial intelligence shows promise in the prediagnosis of oral cancer lesions. The success rates will increase in the training models of the data set that will be formed with more images.

YOLO-V5 based deep learning approach for tooth detection and segmentation on pediatric panoramic radiographs in mixed dentition.

OBJECTIVES: In the interpretation of panoramic radiographs (PRs), the identification and numbering of teeth is an important part of the correct diagnosis. This study evaluates the effectiveness of YOLO-v5 in the automatic detection, segmentation, and numbering of deciduous and permanent teeth in mixed dentition pediatric patients based on PRs. METHODS: A total of 3854 mixed pediatric patients PRs were labelled for deciduous and permanent teeth using the CranioCatch labeling program. The dataset was divided into three subsets: training (n = 3093, 80% of the total), validation (n = 387, 10% of the total) and test (n = 385, 10% of the total). An artificial intelligence (AI) algorithm using YOLO-v5 models were developed. RESULTS: The sensitivity, precision, F-1 score, and mean average precision-0.5 (mAP-0.5) values were 0.99, 0.99, 0.99, and 0.98 respectively, to teeth detection. The sensitivity, precision, F-1 score, and mAP-0.5 values were 0.98, 0.98, 0.98, and 0.98, respectively, to teeth segmentation. CONCLUSIONS: YOLO-v5 based models can have the potential to detect and enable the accurate segmentation of deciduous and permanent teeth using PRs of pediatric patients with mixed dentition.

Atrial myxoma as a mimicker of intracardiac thrombus in Behçet's syndrome: a case study with histopathological confirmation.

Cardiomyopathies cause most intracardiac thrombosis (ICT), and Behçet's syndrome (BS) is a rare inflammatory disease that can be responsible for a proportion of ICT. Other inflammatory disorders involved in the aetiology of ICT include antiphospholipid syndrome, Henoch-Schonlein purpura, COVID-19, and Loeffler endocarditis. ICT usually occur during the active phase of BS, and they have a close relationship with vascular involvement. Atrial myxomas are benign cardiac tumours arising from the interatrial septum. They can lead to a substantial acute phase response, making them difficult to distinguish from inflammatory diseases. In this case study, we present a 46-year-old female BS patient who presented with constitutional symptoms mimicking BS flare in a routine follow-up visit and was diagnosed with left atrial myxoma after administration of several lines of immunosuppressives. Then, she underwent surgical tumour excision, and a histopathological examination confirmed the diagnosis.In conclusion, atrial myxoma should be kept in mind first of all when suspecting ICT, and advanced imaging methods such as cardiac magnetic resonance imaging (MRI) should be used if necessary.

A Deep Learning Approach to Automatic Tooth Caries Segmentation in Panoramic Radiographs of Children in Primary Dentition, Mixed Dentition, and Permanent Dentition.

OBJECTIVES: The purpose of this study was to evaluate the effectiveness of dental caries segmentation on the panoramic radiographs taken from children in primary dentition, mixed dentition, and permanent dentition with Artificial Intelligence (AI) models developed using the deep learning method. METHODS: This study used 6075 panoramic radiographs taken from children aged between 4 and 14 to develop the AI model. The radiographs included in the study were divided into three groups: primary dentition (n: 1857), mixed dentition (n: 1406), and permanent dentition (n: 2812). The U-Net model implemented with PyTorch library was used for the segmentation of caries lesions. A confusion matrix was used to evaluate model performance. RESULTS: In the primary dentition group, the sensitivity, precision, and F1 scores calculated using the confusion matrix were found to be 0.8525, 0.9128, and 0.8816, respectively. In the mixed dentition group, the sensitivity, precision, and F1 scores calculated using the confusion matrix were found to be 0.7377, 0.9192, and 0.8185, respectively. In the permanent dentition group, the sensitivity, precision, and F1 scores calculated using the confusion matrix were found to be 0.8271, 0.9125, and 0.8677, respectively. In the total group including primary, mixed, and permanent dentition, the sensitivity, precision, and F1 scores calculated using the confusion matrix were 0.8269, 0.9123, and 0.8675, respectively. CONCLUSIONS: Deep learning-based AI models are promising tools for the detection and diagnosis of caries in panoramic radiographs taken from children with different dentition.

Microwave-Assisted Hydrothermal Synthesis of Nanosheet Layered Hexagonal Spinel Ferrites for Efficient Electromagnetic Wave Absorbing.

A microwave-induced combustion method has been utilized to prepare nanosized powders of hexagonal spinel ferrites using urea and glycine as fuel. The structural and electromagnetic absorption of these products was investigated using X-ray diffraction (XRD). From the (3,1,1) and (1013) peaks in XRD measurements, it was seen that the samples we synthesized were nanosized spinel and y-phase hexagonal ferrite. In addition, Fourier-transform infrared measurements showed that the synthesized samples had high absorption values showing Fe-O, which is between 878 to 4000 cm-1, and metal-oxygen transitions, which are between the 700 and 2500-3000 cm-1 in the spinel and hexagonal phases. From the scanning electron microscope image of the particles, it is seen that they are homogeneous, filled structures with ∼35-40 nm. It appears from vector network analyzer measurements that the synthesized 1.5 mm thick composite materials have very high electromagnetic absorption properties and that these composites can also be used in electromagnetic interference, Wi-Fi applications, and health.

Evaluation of the relationship between periodontal bone destruction and mesial root concavity of the maxillary first premolar.

BACKGROUND: The purpose of this study was to investigate the morphology of maxillary first premolar mesial root concavity and to analyse its relation to periodontal bone loss (BL) using cone beam computed tomography (CBCT) and panoramic radiographs. METHODS: The mesial root concavity of maxillary premolar teeth was analysed via CBCT. The sex and age of the patients, starting position and depth of the root concavity, apicocoronal length of the concavity on the crown or root starting from the cementoenamel junction (CEJ), total apicocoronal length of the concavity, amount of bone loss both in CBCT images and panoramic radiographs, location of the furcation, length of the buccal and palatinal roots, and buccopalatinal cervical root width were measured. RESULTS: A total of 610 patients' CBCT images were examined, and 100 were included in the study. The total number of upper premolar teeth was 200. The patients were aged between 18 and 65 years, with a mean age of 45.21 ± 13.13 years. All the teeth in the study presented mesial root concavity (100%, n = 200). The starting point of concavity was mostly on the cervical third of the root (58.5%). The mean depth and buccolingual length measurements were 0.96 mm and 4.32 mm, respectively. Depth was significantly related to the amount of alveolar bone loss (F = 5.834, p = 0.001). The highest average concavity depth was 1.29 mm in the group with 50% bone loss. The data indicated a significant relationship between the location of the furcation and bone loss (X2 = 25.215, p = 0.003). Bone loss exceeded 50% in 100% of patients in whom the furcation was in the cervical third and in only 9.5% of patients in whom the furcation was in the apical third (p = 0.003). CONCLUSIONS: According to the results of this study, the depth of the mesial root concavity and the coronal position of the furcation may increase the amount of alveolar bone loss. Clinicians should be aware of these anatomical factors to ensure accurate treatment planning and successful patient management.

Classification of Periapical and Bitewing Radiographs as Periodontally Healthy or Diseased by Deep Learning Algorithms.

Objectives The aim of this artificial intelligence (AI) study was to develop a deep learning algorithm capable of automatically classifying periapical and bitewing radiography images as either periodontally healthy or unhealthy and to assess the algorithm's diagnostic success. Materials and methods The sample of the study consisted of 1120 periapical radiographs (560 periodontally healthy, 560 periodontally unhealthy) and 1498 bitewing radiographs (749 periodontally healthy, 749 periodontally ill). From the main datasets of both radiography types, three sub-datasets were randomly created: a training set (80%), a validation set (10%), and a test set (10%). Using these sub-datasets, a deep learning algorithm was developed with the YOLOv8-cls model (Ultralytics, Los Angeles, California, United States) and trained over 300 epochs. The success of the developed algorithm was evaluated using the confusion matrix method. Results The AI algorithm achieved classification accuracies of 75% or higher for both radiograph types. For bitewing radiographs, the sensitivity, specificity, precision, accuracy, and F1 score values were 0.8243, 0.7162, 0.7439, 0.7703, and 0.7821, respectively. For periapical radiographs, the sensitivity, specificity, precision, accuracy, and F1 score were 0.7500, 0.7500, 0.7500, 0.7500, and 0.7500, respectively. Conclusion The AI models developed in this study demonstrated considerable success in classifying periodontal disease. Future applications may involve employing AI algorithms for assessing periodontal status across various types of radiography images and for automated disease detection.

Synthesis, molecular docking and molecular dynamics simulations, drug-likeness studies, ADMET prediction and biological evaluation of novel pyrazole-carboxamides bearing sulfonamide moiety as potent carbonic anhydrase inhibitors.

Pyrazoles are unique bioactive molecules with a versatile biological profile and they have gained an important place on pharmaceutical chemistry. Pyrazole compounds containing sulfonamide nuclei also attract attention as carbonic anhydrase (CA) inhibitors. In this study, a library of pyrazole-carboxamides were synthesized and the structures of the synthesized molecules were characterized using FT-IR, 1H-NMR, 13C-NMR and HRMS. Then the inhibition effects of newly synthesized molecules on human erythrocyte hCA I and hCA II isoenzymes were investigated. Ki values of the compounds were in the range of 0.063-3.368 µM for hCA I and 0.007-4.235 µM for hCA II. Molecular docking studies were performed between the most active compounds 6a, 6b and the reference inhibitor, acetazolamide (AAZ) and the hCA I and hCA II receptors to investigate the binding mechanisms between the compounds and the receptors. These compounds showed better interactions than the AAZ. ADMET analyzes were performed for the compounds and it was seen that the compounds did not show AMES toxicity. The stability of the molecular docking results over time was analysed by 50 ns molecular dynamics simulations. Molecular dynamics simulations revealed that 6a and 6b exhibited good stability after docking to the binding sites of hCA I and hCA II receptors, with minor conformational changes and fluctuations.

Genomic surveillance during the first two years of the COVID-19 pandemic - country experience and lessons learned from Türkiye.

Background: Türkiye confirmed its first case of SARS-CoV-2 on March 11, 2020, coinciding with the declaration of the global COVID-19 pandemic. Subsequently, Türkiye swiftly increased testing capacity and implemented genomic sequencing in 2020. This paper describes Türkiye's journey of establishing genomic surveillance as a middle-income country with limited prior sequencing capacity and analyses sequencing data from the first two years of the pandemic. We highlight the achievements and challenges experienced and distill globally relevant lessons. Methods: We tracked the evolution of the COVID-19 pandemic in Türkiye from December 2020 to February 2022 through a timeline and analysed epidemiological, vaccination, and testing data. To investigate the phylodynamic and phylogeographic aspects of SARS-CoV-2, we used Nextstrain to analyze 31,629 high-quality genomes sampled from seven regions nationwide. Results: Türkiye's epidemiological curve, mirroring global trends, featured four distinct waves, each coinciding with the emergence and spread of variants of concern (VOCs). Utilizing locally manufactured kits to expand testing capacity and introducing variant-specific quantitative reverse transcription polymerase chain reaction (RT-qPCR) tests developed in partnership with a private company was a strategic advantage in Türkiye, given the scarcity and fragmented global supply chain early in the pandemic. Türkiye contributed more than 86,000 genomic sequences to global databases by February 2022, ensuring that Turkish data was reflected globally. The synergy of variant-specific RT-qPCR kits and genomic sequencing enabled cost-effective monitoring of VOCs. However, data analysis was constrained by a weak sequencing sampling strategy and fragmented data management systems, limiting the application of sequencing data to guide the public health response. Phylodynamic analysis indicated that Türkiye's geographical position as an international travel hub influenced both national and global transmission of each VOC despite travel restrictions. Conclusion: This paper provides valuable insights into the testing and genomic surveillance systems adopted by Türkiye during the COVID-19 pandemic, proposing important lessons for countries developing national systems. The findings underscore the need for robust testing and sampling strategies, streamlined sample referral, and integrated data management with metadata linkage and data quality crucial for impactful epidemiological analysis. We recommend developing national genomic surveillance strategies to guide sustainable and integrated expansion of capacities built for COVID-19 and to optimize the effective utilization of sequencing data for public health action.

The Detection of Pulp Stones with Automatic Deep Learning in Panoramic Radiographies: An AI Pilot Study.

This study aims to evaluate the effectiveness of employing a deep learning approach for the automated detection of pulp stones in panoramic imaging. A comprehensive dataset comprising 2409 panoramic radiography images (7564 labels) underwent labeling using the CranioCatch labeling program, developed in Eskisehir, Turkey. The dataset was stratified into three distinct subsets: training (n = 1929, 80% of the total), validation (n = 240, 10% of the total), and test (n = 240, 10% of the total) sets. To optimize the visual clarity of labeled regions, a 3 × 3 clash operation was applied to the images. The YOLOv5 architecture was employed for artificial intelligence modeling, yielding F1, sensitivity, and precision metrics of 0.7892, 0.8026, and 0.7762, respectively, during the evaluation of the test dataset. Among deep learning-based artificial intelligence algorithms applied to panoramic radiographs, the use of numerical identification for the detection of pulp stones has achieved remarkable success. It is expected that the success rates of training models will increase by using datasets consisting of a larger number of images. The use of artificial intelligence-supported clinical decision support system software has the potential to increase the efficiency and effectiveness of dentists.

Tooth numbering and classification on bitewing radiographs: an artificial intelligence pilot study.

OBJECTIVE: The aim of this study is to assess the efficacy of employing a deep learning methodology for the automated identification and enumeration of permanent teeth in bitewing radiographs. The experimental procedures and techniques employed in this study are described in the following section. STUDY DESIGN: A total of 1248 bitewing radiography images were annotated using the CranioCatch labeling program, developed in Eskisehir, Turkey. The dataset has been partitioned into 3 subsets: training (n = 1000, 80% of the total), validation (n = 124, 10% of the total), and test (n = 124, 10% of the total) sets. The images were subjected to a 3 × 3 clash operation in order to enhance the clarity of the labeled regions. RESULTS: The F1, sensitivity and precision results of the artificial intelligence model obtained using the Yolov5 architecture in the test dataset were found to be 0.9913, 0.9954, and 0.9873, respectively. CONCLUSION: The utilization of numerical identification for teeth within deep learning-based artificial intelligence algorithms applied to bitewing radiographs has demonstrated notable efficacy. The utilization of clinical decision support system software, which is augmented by artificial intelligence, has the potential to enhance the efficiency and effectiveness of dental practitioners.

A 25-Year-Old Man with a History of Substance Abuse Presenting with Pneumomediastinum Due to Methamphetamine Vapor Inhalation.

BACKGROUND There has been an increase in the use of inhalation methods to abuse drugs, including freebasing crack cocaine (alkaloid) and inhaling methamphetamine vapor. This report is of a 25-year-old man with a history of substance abuse presenting with pneumomediastinum due to methamphetamine vapor inhalation. Acute pneumomediastinum is an extremely rare complication of methamphetamine use. CASE REPORT A 25-year-old man was treated for polysubstance abuse following 9 days of methamphetamine abuse. EKG did not show any ST &T change. D-dimer was normal, at 0.4 mg/L, so we did not do further work-up for pulmonary embolism. His chest pain worsened in the Emergency Department (ED), and a physical exam demonstrated crepitation of the posterior neck, trapezius, and right scapula. A portable chest X-ray revealed subcutaneous air over the right scapular region, in addition to pneumomediastinum. The urine drug screen test was positive for methamphetamine. A chest CT was ordered, which showed a moderate-volume pneumomediastinum with soft-tissue air tracking into the lower neck and along the right chest wall. The patient underwent an esophagogram, which showed no air leak, and Boerhaave's syndrome was ruled out. His symptoms improved and he did not require any surgical intervention. CONCLUSIONS Considering the higher rates of illicit substance use, especially methamphetamine, it is important to pay attention to the associated pathologies and to keep spontaneous pneumomediastinum on the list of differentials for patients using methamphetamine, particularly those who inhale it, which can cause pneumomediastinum, even without Boerhaave's syndrome.

Detecting white spot lesions on post-orthodontic oral photographs using deep learning based on the YOLOv5x algorithm: a pilot study.

BACKGROUND: Deep learning model trained on a large image dataset, can be used to detect and discriminate targets with similar but not identical appearances. The aim of this study is to evaluate the post-training performance of the CNN-based YOLOv5x algorithm in the detection of white spot lesions in post-orthodontic oral photographs using the limited data available and to make a preliminary study for fully automated models that can be clinically integrated in the future. METHODS: A total of 435 images in JPG format were uploaded into the CranioCatch labeling software and labeled white spot lesions. The labeled images were resized to 640 × 320 while maintaining their aspect ratio before model training. The labeled images were randomly divided into three groups (Training:349 images (1589 labels), Validation:43 images (181 labels), Test:43 images (215 labels)). YOLOv5x algorithm was used to perform deep learning. The segmentation performance of the tested model was visualized and analyzed using ROC analysis and a confusion matrix. True Positive (TP), False Positive (FP), and False Negative (FN) values were determined. RESULTS: Among the test group images, there were 133 TPs, 36 FPs, and 82 FNs. The model's performance metrics include precision, recall, and F1 score values of detecting white spot lesions were 0.786, 0.618, and 0.692. The AUC value obtained from the ROC analysis was 0.712. The mAP value obtained from the Precision-Recall curve graph was 0.425. CONCLUSIONS: The model's accuracy and sensitivity in detecting white spot lesions remained lower than expected for practical application, but is a promising and acceptable detection rate compared to previous study. The current study provides a preliminary insight to further improved by increasing the dataset for training, and applying modifications to the deep learning algorithm. CLINICAL REVELANCE: Deep learning systems can help clinicians to distinguish white spot lesions that may be missed during visual inspection.