Article
Chemistry, Multidisciplinary
Jood Bazerbashi, Qoot Alkhubaizi, Azin Parsa, Mohamed Shabayek, Howard Strassler, Mary Anne S. Melo
Summary: The prevalence, location, and characteristics of radiolucency findings associated with proximal class II composite restorations were studied. The majority (83.5%) of the examined bitewing radiographs showed unusual radiographic signs, including internal voids, overhang, interlayer lines, secondary caries, interfacial gaps, and multiple atypical findings. This highlights the challenge of diagnosing radiolucencies underneath composite restorations.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Analytical
Maira Moran, Marcelo Faria, Gilson Giraldi, Luciana Bastos, Larissa Oliveira, Aura Conci
Summary: This study proposes a method that combines image processing techniques and convolutional neural networks to identify approximal dental caries in bitewing radiographic images and classify them according to lesion severity. Using 112 radiographs, individual tooth images were extracted, labeled, and used to train CNN models. Evaluation was performed using different learning rates and architectures, with the Inception model achieving the best accuracy of 73.3% at a learning rate of 0.001. The results suggest that the proposed method could be useful in assisting dentists in evaluating bitewing images and defining lesion severity for appropriate treatments.
Article
Medicine, General & Internal
Gian Andrea Pelliccioni, Maria Rosaria Antonella Gatto, Silvia Bolognesi, Daniele Dal Fiume, Maicon Sebold, Lorenzo Breschi
Summary: The study evaluated the accuracy and efficiency of DIFOTI in detecting approximal caries in posterior teeth compared to intra-oral examination and bitewing radiographs. Results showed that DIFOTI was accurate and faster than traditional methods, highlighting its clinical relevance for caries diagnosis.
JOURNAL OF CLINICAL MEDICINE
(2021)
Article
Multidisciplinary Sciences
Shinae Lee, Sang-il Oh, Junik Jo, Sumi Kang, Yooseok Shin, Jeong-won Park
Summary: The study developed a CNN model using U-Net for caries detection on bitewing radiographs, and demonstrated that the model can help clinicians diagnose caries more accurately, especially in cases of initial and moderate caries, as shown by the improved diagnostic performance of three dentists using the model's results.
SCIENTIFIC REPORTS
(2021)
Article
Dentistry, Oral Surgery & Medicine
Yusuf Bayraktar, Enes Ayan
Summary: The study demonstrates that deep convolutional neural networks show good performance in diagnosing caries lesions in digital bitewing radiographs, providing assistance for accurate diagnosis and treatment.
CLINICAL ORAL INVESTIGATIONS
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Wannakamon Panyarak, Kittichai Wantanajittikul, Arnon Charuakkra, Sangsom Prapayasatok, Wattanapong Suttapak
Summary: This study aimed to evaluate the impact of image size, IoU thresholds, and confidence thresholds on the performance of YOLO models in dental caries detection. The results showed that YOLOv7 outperformed YOLOv3 in terms of precision, F1-score, and mAP, while having a lower recall. Increasing the image size did not enhance the model's performance, and changing IoU and confidence thresholds had mixed effects on the model's performance.
JOURNAL OF DIGITAL IMAGING
(2023)
Article
Dentistry, Oral Surgery & Medicine
Jingwei Cao, Yuwen Fang, Yue Liao, Yan Wang, Ran Yang, Yang Zhang, Qiong Zhang, Jing Zou
Summary: NIRI showed higher sensitivity and lower specificity compared with VI in detecting proximal caries in primary molars. Therefore, it is recommended to use NIRI in combination with BWR to improve the detection rate of proximal caries in primary molars.
JOURNAL OF DENTISTRY
(2023)
Article
Dentistry, Oral Surgery & Medicine
Deivi Cascante-Sequeira, Hugo Gaeta-Araujo, Danieli Moura Brasil, Deborah Queiroz Freitas, Francisco Haiter-Neto
Summary: This study aimed to assess the reproducibility of a wedge-guided bitewing image receptor-holding device (IRHD-WG) compared to a commercially available bitewing image receptor-holding device (IRHD-XCP). The results showed that IRHD-WG had higher reproducibility, with more similar radiographs and better diagnostic value.
DENTOMAXILLOFACIAL RADIOLOGY
(2022)
Article
Dentistry, Oral Surgery & Medicine
Zvi Metzger, Dana G. Colson, Peggy Bown, Timo Weihard, Ingo Baresel, Tim Nolting
Summary: The study compared the effectiveness of near-infrared light reflection (NILR) and bitewing radiography (BWR) in detecting proximal caries. NILR was found to be more sensitive in detecting early enamel lesions compared to BWR, and comparable in detecting lesions involving the dentino-enamel junction (DEJ). Expert team showed higher accuracy in caries detection using both methods.
JOURNAL OF DENTISTRY
(2022)
Article
Chemistry, Analytical
Song Hee Oh, Sae Rom Lee, Jin Young Choi, Yong Suk Choi, Seong Hun Kim, Hong Cheol Yoon, Gerald Nelson
Summary: By comparing and analyzing conventional examination with the QLF technique, this study aimed to present an optimal diagnostic protocol. It was found that QLF showed higher sensitivity in detecting occlusal dental caries and cracks compared to the conventional method. The QLF technique may be a useful adjunct tool for the detection of occlusal caries and peripheral cracks.
Article
Dentistry, Oral Surgery & Medicine
Angela Dalla Nora, Luana Severo Alves, Nathalia Costa de Castro, Marisa Maltz, Julio Eduardo do Amaral Zenkner
Summary: This study evaluated the radiographic features of inactive enamel caries lesions in permanent molars and found that lesions with radiolucency at baseline were more likely to progress over 4-5 years. The presence of radiolucency at baseline was a predictor of caries progression, indicating the importance of closer monitoring for such lesions. The use of bitewing radiographs can help identify surfaces/lesions more likely to progress and establish proper recall intervals for patient monitoring.
JOURNAL OF DENTISTRY
(2021)
Article
Computer Science, Hardware & Architecture
G. Vimalarani, Uppu Ramachandraiah
Summary: In this study, a dental radiography technique based on image processing and neural network approaches was proposed for the accurate identification and classification of dental caries. By preprocessing the input images and extracting features using deep learning methods, the caries can be successfully detected and classified with the use of a deep gradient-based LeNet classifier model.
MICROPROCESSORS AND MICROSYSTEMS
(2022)
Article
Dentistry, Oral Surgery & Medicine
Lukas Kunt, Jan Kybic, Valeria Nagyova, Antonin Tichy
Summary: The objective of this study was to create an annotated dataset of bitewing radiographs and use convolutional neural networks to automate the detection of dental caries. The study found that by using model ensembling and post-processing, the detection performance could be improved to achieve accuracy comparable to experienced dentists.
CLINICAL ORAL INVESTIGATIONS
(2023)
Review
Dentistry, Oral Surgery & Medicine
Ann Wenzel
Summary: Radiographic imaging for caries diagnosis has a long history and has been continuously evolving. Various methods and X-ray receptors have been developed, while computer systems have made progress in aiding dentists in detecting lesions and estimating depth.
DENTOMAXILLOFACIAL RADIOLOGY
(2021)
Article
Chemistry, Multidisciplinary
Vincent Majanga, Serestina Viriri
Summary: Dental caries are a prevalent chronic disease globally, and early detection is crucial for treatment. Traditional computer aided diagnosis methods are limited by the complex characteristics of caries images, unable to effectively identify hidden lesions. The proposed deep learning technique utilizes blob detection for automatic detection of caries in dental radiographs.
APPLIED SCIENCES-BASEL
(2021)
Article
Dentistry, Oral Surgery & Medicine
Michael Naumann, Patricia Scholz, Joachim Krois, Falk Schwendicke, Guido Sterzenbach, Arndt Happe
Summary: The objective of this study was to compare monolithic hybrid abutment crowns (screw-retained) versus monolithic hybrid abutments with adhesively cemented monolithic single-tooth crowns for implant-borne restorations. Twenty subjects were randomly assigned to receive either cement-retained or screw-retained monolithic restorations. The follow-up period was 36 months, and the results showed no significant difference between the two treatment options in terms of patient satisfaction and restoration outcomes.
CLINICAL ORAL IMPLANTS RESEARCH
(2023)
Review
Dentistry, Oral Surgery & Medicine
Kuo Feng Hung, Andy Wai Kan Yeung, Michael M. Bornstein, Falk Schwendicke
Summary: Personalized medicine tailors diagnostics and therapeutics to individuals based on their biological, social, and behavioral characteristics. Advanced AI technologies have the potential to integrate diverse data and bring us closer to personalized dentistry, also known as P4 dentistry. AI applications in dentomaxillofacial imaging have shown promising results, but further research is needed to assess their impact on treatment decisions and outcomes.
DENTOMAXILLOFACIAL RADIOLOGY
(2023)
Article
Dentistry, Oral Surgery & Medicine
S. R. Herbst
Summary: This study aimed to identify significant associations between preoperative risk factors and achieving optimal root filling length (RFL) during orthograde root canal treatments (RCT) and to predict successful RFL using machine learning. The results showed that the success of RFL during RCT is influenced by the operator and several risk factors. However, the predictive performance of machine learning algorithms for RFL was insufficient.
JOURNAL OF DENTISTRY
(2023)
Review
Medicine, General & Internal
Lubaina T. Arsiwala-Scheppach, Akhilanand Chaurasia, Anne Mueller, Joachim Krois, Falk Schwendicke
Summary: Machine learning is increasingly used in dental research and application. This study aimed to compile and evaluate studies using machine learning in dentistry, focusing on methodological quality and reporting standards. The findings showed that machine learning is commonly employed for classification tasks in different clinical fields. Forty-two different metrics were used to evaluate model performances, with accuracy, sensitivity, precision, and intersection-over-union being the most common. The study identified a significant risk of bias and moderate adherence to reporting standards, highlighting the need for a minimum (core) set of outcome and outcome metrics for comparison across studies.
JOURNAL OF CLINICAL MEDICINE
(2023)
Review
Medicine, General & Internal
Marcus Cebula, Gerd Goestemeyer, Joachim Krois, Vinay Pitchika, Sebastian Paris, Falk Schwendicke, Susanne Effenberger
Summary: The present study conducted a meta-analysis to evaluate the certainty of evidence for resin infiltration of proximal carious lesions in primary and permanent teeth. The results showed that resin infiltration can effectively arrest caries lesions in both primary and permanent teeth. However, the certainty of evidence remains unclear due to the high or unclear risk of bias in the included trials.
JOURNAL OF CLINICAL MEDICINE
(2023)
Article
Immunology
Stella Danek, Martha Buettner, Joachim Krois, Falk Schwendicke
Summary: To reach more vaccine recipients, high-income countries have introduced mass vaccination centers for COVID-19. Understanding user experiences of these structures can help improve their design and increase patient satisfaction and vaccine uptake.
Article
Biology
Martin Baxmann, Lan Huong Timm, Falk Schwendicke
Summary: The key step prior to clear aligner therapy (CAT) is to examine and select suitable cases, which includes understanding the orthodontic problem and evaluating oral health. The characteristics of patients seeking CAT may differ among European countries. This study retrospectively analyzed anonymized data from a large CAT provider in five European countries. The results showed differences in gender distribution and motivations across countries, as well as low prevalence of oral conditions among patients seeking CAT.
Editorial Material
Dentistry, Oral Surgery & Medicine
Martha Buettner, Falk Schwendicke
BRITISH DENTAL JOURNAL
(2023)
Article
Dentistry, Oral Surgery & Medicine
W. Park, F. Schwendicke, J. Krois, J. -K. Huh, J. -H. Lee
Summary: This study evaluated the efficacy of deep learning in identifying and classifying different types of dental implant systems using a large-scale multicenter data set. Deep learning outperformed dental professionals in accuracy and required less reading and classification time. The results suggest that deep learning can be used successfully as a decision support aid for the identification and classification of dental implant systems encountered in clinical practice.
JOURNAL OF DENTAL RESEARCH
(2023)
Review
Dentistry, Oral Surgery & Medicine
Hossein Mohammad-Rahimi, Rata Rokhshad, Sompop Bencharit, Joachim Krois, Falk Schwendicke
Summary: The objective of this study is to explain the basic concept of deep learning to dental researchers and professionals who may find it challenging to understand and interpret deep learning studies. The study provides an overview of deep learning models, approaches, and data management strategies, as well as a step-by-step guide for completing a real-world project.
JOURNAL OF DENTISTRY
(2023)
Article
Dentistry, Oral Surgery & Medicine
Lisa Schneider, Roman Rischke, Joachim Krois, Aleksander Krasowski, Martha Buettner, Hossein Mohammad-Rahimi, Akhilanand Chaurasia, Nielsen S. Pereira, Jae-Hong Lee, Sergio E. Uribe, Shahriar Shahab, Revan Birke Koca-Unsal, Gurkan Unsal, Yolanda Martinez-Beneyto, Janet Brinz, Olga Tryfonos, Falk Schwendicke
Summary: This study demonstrates the effectiveness of Federated Learning (FL) in dentistry, showing that it can train high-performance and generalizable deep learning models without sharing sensitive data. FL outperformed Local Learning (LL) in tooth segmentation on panoramic radiographs and proved to be a valuable alternative when data pooling is not feasible.
JOURNAL OF DENTISTRY
(2023)
Article
Dentistry, Oral Surgery & Medicine
Falk Schwendicke, Lisa Bombeck
Summary: This study assessed the cost-effectiveness of near-infrared-light-transillumination (NILT) for school-based caries screening and found that NILT may provide limited effectiveness gains and cost savings in the modelled populations. However, in countries other than Germany, NILT may have higher cost-effectiveness.
JOURNAL OF DENTISTRY
(2023)
Article
Engineering, Biomedical
Philipp Messer-Hannemann, Henrik Bottcher, Sven Henning, Falk Schwendicke, Susanne Effenberger
Summary: The objective of this study was to apply the concept of ductile particle reinforcement to restorative dentistry and introduce an innovative glass ionomer material based on the dispersion of PEG-PU micelles. The results showed that reinforcing a conventional glass ionomer with PEG-PU micelles significantly improved the flexural strength and fracture toughness, enhancing the mechanical properties of the material.
JOURNAL OF FUNCTIONAL BIOMATERIALS
(2023)
Article
Dentistry, Oral Surgery & Medicine
Gerd Goestemeyer, Mareike Preus, Karim Elhennawy, Falk Schwendicke, Sebastian Paris, Haitham Askar
Summary: In this study, the diagnostic accuracy of radiographic evaluation, visual-tactile assessment, laser-fluorescence, and near-infrared-light transillumination were evaluated for proximal root caries lesions. The results showed that near-infrared-light transillumination was not applicable for detecting proximal root caries, while laser-fluorescence and visual-tactile assessment were more accurate than radiographic evaluation for detecting these lesions.
CLINICAL ORAL INVESTIGATIONS
(2023)
Article
Dentistry, Oral Surgery & Medicine
Chang Zeng, Pei Hu, Colin P. Egan, Brian E. Bergeron, Franklin Tay, Jingzhi Ma
Summary: The study compared the bacteria debridement efficacy of the first and second generations of sonic root canal irrigant activation systems. The results showed that activation of 2% NaOCl with either EndoActivator or SmartLite Pro EndoActivator significantly reduced the overall intracanal bacterial load compared to syringe-side-vented needle delivery. However, delivering 2% NaOCl for 5 minutes using the syringe-side-vented needle produced better bacteria debridement than either sonic agitation system.
JOURNAL OF DENTISTRY
(2024)
Article
Dentistry, Oral Surgery & Medicine
Alaa Husni Qari, Raghad Mohammed Alharbi, Shahd Saud Alomiri, Banan Nasser Alandanusi, Lina Ayman Mirza, Mohammad Hasan Al-Harthy
Summary: The study compares the experience of patients seen using teledentistry with traditional Orofacial Pain and TMD visits. Results indicated a borderline difference in effectively using teledentistry for follow-ups among different age groups, but no significant differences were found in patients' experience between virtual and conventional visits. This suggests that high-quality dental care services can be provided remotely without adversely affecting patient experience or quality of care.
JOURNAL OF DENTISTRY
(2024)
Article
Dentistry, Oral Surgery & Medicine
Giovanna Iezzi, Barbara Zavan, Morena Petrini, Letizia Ferroni, Tania Vanessa Pierfelice, Ugo D'Amora, Alfredo Ronca, Emira D'Amico, Carlo Mangano
Summary: This study characterized the surface topography and evaluated the biological responses of various cells to a novel 3D-printed dental implant. The surface of the implant showed a highly interconnected porous architecture and roughness, leading to good adhesion and growth of human oral osteoblasts, gingival fibroblasts, mesenchymal stem cells, and monocytes. These findings suggest that this 3D-printed implant has excellent potential for clinical applications.
JOURNAL OF DENTISTRY
(2024)