Dento maxillo facial radiology最新文献

Objectives: Accurate segmentation of the maxillary sinus from medical images is essential for diagnostic purposes and surgical planning. Manual segmentation of the maxillary sinus, while the gold standard, is time consuming and requires adequate training. To overcome this problem, artificial intelligence (AI) enabled automatic segmentation software's developed. The purpose of this review is to systematically analyse the current literature to investigate the accuracy and efficiency of automatic segmentation techniques of the maxillary sinus to manual segmentation.

Methods: A systematic approach to perform a thorough analysis of the existing literature using PRISMA guidelines. Data for this study was obtained from Pubmed, Medline, Embase, and Google Scholar databases. The inclusion and exclusion eligibility criteria were used to shortlist relevant studies. The sample size, anatomical structures segmented, experience of operators, type of manual segmentation software used, type of automatic segmentation software used, statistical comparative method used, and length of time of segmentation were analysed.

Results: This systematic review presents 10 studies that compared the accuracy and efficiency of automatic segmentation of the maxillary sinus to manual segmentation. All the studies included in this study were found to have a low risk of bias. Samples sizes ranged from 3 to 144, a variety of operators were used to manually segment the cone-beam computed tomography (CBCT) and segmentation was made primarily to 3D slicer and Mimics software. The comparison was primarily made to Unet architecture softwares, with the dice-coefficient being the primary means of comparison.

Conclusions: This systematic review showed that automatic segmentation technique was consistently faster than manual segmentation techniques and over 90% accurate when compared to the gold standard of manual segmentation.

Objectives: To compare the accuracy of cone-beam CT (CBCT), ultrasonography (USG) and direct measurements in linear dimensions of periodontal defects on the buccal alveolar surfaces of mandibular sheep teeth.

Methods: A total of 88 defects were artificially created, including dehiscence, fenestration, grade I and II endodontic-periodontal defects. Two observers performed measurements twice. Maximum length, depth, and width of the defects were measured with all 3 methods. Manual measurements were accepted as the gold standard. Intraclass correlation coefficients (ICC) were calculated. The mean value of the measurements, the bias, the SD of the differences, and the limits of agreement were estimated. Statistical significance was set at P < .05.

Results: Intra- and inter-observer reliability was excellent, suggesting ICCs 0.988-1 and 0.981-1, respectively. The highest CCs were obtained from depth measurements, while the lowest CCs were obtained from length measurements. Although the differences were scattered around the bias. The estimated bias values for USG and CBCT were 0.18 (0.153-0.21) (P < .001) and 0.091 (0.079-0.102) (P < .001), respectively. Observers recorded measurements which were slightly underestimated with both techniques utilized.

Conclusions: Observers measured periodontal defects with clinically acceptable underestimations by using CBCT and USG.

Advances in knowledge: It is important to compare different innovative imaging modalities and gauge their efficiency in the measurement of various types of periodontal defects in terms of treatment planning, prognosis, and follow up of those cases.

Objectives: Due to the difference between the natural head position during scan and the orientation of CBCT display required for diagnosis, radiologists need to manually adjust the image orientation during clinical diagnosis. To eliminate this difference, this study explored orientation normalization algorithm for mandibular condyle in the small field-of-view (FoV) cone beam CT (CBCT) images.

Methods: Based on the morphology analysis, we designed principal component analysis (PCA) based orientation normalization algorithm for condyle in the small FoV CBCT images. The algorithm involves first locating the reference centre, defined as the centre coordinates of the condylar head in the maximum axial plane, through segmentation and centroid calculation. Subsequently, the maximum principal orientations in the axial, coronal, and sagittal planes are extracted using PCA algorithm. Finally, the condyle orientation is normalized by using rotation transformation matrices derived from condylar head centre localization and principal orientation extraction.

Results: Our algorithm was evaluated on 2 CBCT image datasets with 692 scans, and multiple experiments were designed from aspects of algorithm accuracy and stability. Experimental results demonstrate that images with orientation normalization are consistent with the radiologists expected perspective from both qualitative and quantitative aspects. The normalized results of CBCT images taken at multiple time-points also further confirm that our method has good stability.

Conclusion: Based on the morphological characteristics, medical image processing algorithm can achieve accurate and stable orientation normalization for condyle in the small FoV CBCT images.

Objectives: Deep learning-driven super resolution (SR) aims to enhance the quality and resolution of images, offering potential benefits in dental imaging. Although extensive research has focused on deep learning based dental classification tasks, the impact of applying SR techniques on classification remains underexplored. This study seeks to address this gap by evaluating and comparing the performance of deep learning classification models on dental images with and without SR enhancement.

Methods: An open-source dental image dataset was utilized to investigate the impact of SR on image classification performance. SR was applied by 2 models with a scaling ratio of 2 and 4, while classification was performed by 4 deep learning models. Performances were evaluated by well-accepted metrics like structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), accuracy, recall, precision, and F1 score. The effect of SR on classification performance is interpreted through 2 different approaches.

Results: Two SR models yielded average SSIM and PSNR values of 0.904 and 36.71 for increasing resolution with 2 scaling ratios. Average accuracy and F-1 score for the classification trained and tested with 2 SR-generated images were 0.859 and 0.873. In the first of the comparisons carried out with 2 different approaches, it was observed that the accuracy increased in at least half of the cases (8 out of 16) when different models and scaling ratios were considered, while in the second approach, SR showed a significantly higher performance for almost all cases (12 out of 16).

Conclusion: This study demonstrated that the classification with SR-generated images significantly improved outcomes.

Advances in knowledge: For the first time, the classification performance of dental radiographs with improved resolution by SR has been investigated. Significant performance improvement was observed compared to the case without SR.

Objectives: The present study examined the potential use of CT panoramic mandibular indices on cone beam CT (CBCT) for the assessment of bone density in patients with primary and secondary causes of low bone mass.

Study design: The study enrolled 104 postmenopausal women and 66 patients with endocrine-related low bone mass (diabetes mellitus, acromegaly, Cushing syndrome), who underwent dual-energy X-ray absorptiometry (DXA) and CBCT scanning. The study assessed the correlation between DXA parameters (lumbar spine, femoral neck, total hip T-score, bone mineral density [BMD], and trabecular bone score [TBS]) and CBCT-derived indices (CT mandibular index superior [CTI(S)], CT mandibular index inferior [CTI(I)], and CT mental index [CTMI]).

Results: Significant correlations were found between the CBCT indices and both quantitative (BMD, T-score) and qualitative (TBS) measures of bone mass. In postmenopausal women, all 3 CBCT indices showed strong correlations with DXA parameters. In secondary endocrine causes, CTMI and CTI(S) were significantly correlated with TBS scores, and CTMI also showed a significant correlation with lumbar BMD.

Conclusion: The study concludes that CTI(S), CTI(I), and CTMI are valuable for assessing bone density and quality in patients with low bone mass, both in primary and secondary osteoporosis related to diabetes mellitus, acromegaly, and Cushing syndrome.

Advances in knowledge: These findings support the use of CBCT as a useful tool for evaluating bone health in the clinical setting and optimizing dental implant result. It is among the first studies to evaluate bone mass quality and quantity in secondary endocrine causes of low bone mass.

Objectives: To conduct a prospective cohort study evaluating the influence of different teaching methodologies on the radiographic diagnosis of root resorptions by undergraduate dental students.

Methods: Forty-eight undergraduate students were randomly divided into 4 groups (n = 12) according to the methodology applied to teach about root resorption: traditional face-to-face teaching (control), remote teaching, gamification, and case study. The first stage was to perform a pre-methodology index test to assess prior knowledge about root resorption. Then, all groups received study material on a virtual platform and 1 week later, the teaching methodologies were applied. Twenty-four hours after each methodology application, the students performed a diagnostic test by analysing 28 digital periapical radiographs, classifying them according to the absence or type of root resorption present (external superficial, internal inflammatory, or external cervical). After 10 days, 3 students in each group (25%) were randomly selected and re-evaluated the 28 images to calculate intra-rater agreement. All students repeated the index test 30 days after the interventions. Statistical analysis used linear regression models, Pearson's correlation, and chi-square test (P < .05).

Results: Gamification resulted in better student performance in the index and radiographic diagnostic tests (P < .001). Superficial external resorption was the most challenging to diagnose, regardless of the method, while inflammatory internal obtained a higher percentage of correct responses (P < .001) in the diagnostic test.

Conclusions: All methods involving student interaction demonstrated better outcomes compared to the traditional model in the diagnosis of root resorptions. Gamification resulted in the best performance and may be an effective resource in the learning process.

Advancements in knowledge: Adopting gamification enhanced student performance and may be a valuable learning strategy to contribute to a more accurate diagnosis and safer clinical practice.

Objectives: To evaluate patients' perceptions of the use of artificial intelligence (AI) in dental imaging diagnostics across 6 centres worldwide, hereby named according to their respective cities: Ribeirão Preto (Brazil), Aarhus (Denmark), Lyon (France), Tromsø (Norway), Porto (Portugal), Louisville (USA).

Methods: A survey was administered at each centre, focusing on patient attitudes and beliefs regarding AI in dental imaging diagnostics. The survey comprised 16 statements rated on a Likert scale, patient characteristics, and an optional comment section. Inter-centre differences were analysed using chi-square and Fisher's exact tests, and correlation analyses were performed between participant characteristics and their perceptions of AI.

Results: A total of 2,581 responses were collected. Most participants expressed positive perceptions of AI as a complementary diagnostic tool, rather than a replacement for human dentists. Key concerns included the need for human oversight, data privacy, and potential cost increases. Differences were observed between centres, with participants from Ribeirão Preto being more likely to accept AI replacing dentists, whereas those from Aarhus and Tromsø expressed greater scepticism about AI's diagnostic capabilities. Although higher levels of education and knowledge about AI correlated with more optimistic perspectives about AI's diagnostic capabilities, they were also associated with an increased preference for human supervision.

Conclusions: Overall, patients favour the use of AI in dental imaging as an auxiliary diagnostic tool, with human supervision remaining essential. Cultural and demographic factors significantly influence perceptions.

Advances in knowledge: The findings highlight the need for tailored communication strategies to address patient concerns if AI is integrated into dental care.

Objectives: This study explores the application of artificial intelligence (AI), specifically deep learning, in the detection and classification of mandibular fractures using CT scans.

Methods: Data from 459 patients were retrospectively obtained from West China Hospital of Stomatology, Sichuan University, spanning from 2020 to 2023. The CT scans were divided into training, testing, and independent validation sets. This research focuses on training and validating a deep learning model using the nnU-Net segmentation framework for pixel-level accuracy in identifying fracture locations. Additionally, a 3D-ResNet with pre-trained weights was employed to classify fractures into 3 types based on severity. Performance metrics included sensitivity, precision, specificity, and area under the receiver operating characteristic curve (AUC).

Results: The study achieved high diagnostic accuracy in mandibule fracture detection, with sensitivity >0.93, precision >0.79, and specificity >0.80. For mandibular fracture classification, accuracies were all above 0.718, with a mean AUC of 0.86.

Conclusions: Detection and classification of mandibular fractures in CT images can be significantly enhanced using the nnU-Net segmentation framework, aiding in clinical diagnosis.

Objectives: To evaluate deep learning (DL)-based models for detecting periapical bone rarefaction (PBRs) in panoramic radiographs (PRs), analysing their feasibility and performance in dental practice.

Methods: A search was conducted across seven databases and partial grey literature up to November 15, 2024, using Medical Subject Headings and entry terms related to DL, PBRs, and PRs. Studies assessing DL-based models for detecting and classifying PBRs in conventional PRs were included, while those using non-PR imaging or focusing solely on non-PBR lesions were excluded. Two independent reviewers performed screening, data extraction, and quality assessment using the Quality Assessment of Diagnostic Accuracy Studies-2 tool, with conflicts resolved by a third reviewer.

Results: Twelve studies met the inclusion criteria, mostly from Asia (58.3%). The risk of bias was moderate in 10 studies (83.3%) and high in 2 (16.7%). DL models showed moderate to high performance in PBR detection (sensitivity: 26%-100%; specificity: 51%-100%), with U-NET and YOLO being the most used algorithms. Only one study (8.3%) distinguished Periapical Granuloma from Periapical Cysts, revealing a classification gap. Key challenges included limited generalization due to small datasets, anatomical superimpositions in PRs, and variability in reported metrics, compromising models comparison.

Conclusion: This review underscores that DL-based has the potential to become a valuable tool in dental image diagnostics, but it cannot yet be considered a definitive practice. Multicentre collaboration is needed to diversify data and democratize those tools. Standardized performance reporting is critical for fair comparability between different models.

Advances in knowledge: This study represents the first critical synthesis on this theme, examining a group of lesions with complex manifestations that have been neglected in comparable technological development studies, where research focus has usually been limited to radicular cysts. We identified gaps in classification tasks, insufficient use of ethnically diverse and heterogeneous datasets, and the need for multicentric studies. The variability in data reporting prevents transparent comparisons, even precluding our planned meta-analysis. Consequently, we emphasize the necessity for standardized reporting protocols similar to PRISMA for systematic reviews or STARD for diagnostic or prognostic studies, particularly since accuracy metrics remain inadequately documented while critically important.

Digital X-ray sensors have significantly changed dental radiography, enabling faster image acquisition and reducing radiation doses for patients. Despite the advancements in technology, noise in X-ray imaging remains a challenge. In this study, noise was examined using the Noise-Power Spectrum (NPS) and a non-parametric method. Blank images were taken under different exposure times and voltage settings. The analyses show that noise decreases with longer exposure times. Among the examined sensors, 2 showed distinct NPS peaks, and 1 exhibited no relationship between exposure time and noise levels. These results are discussed on terms of specific sensor structures, artefacts and/or unaccessible post-processing algorithms.