Dento maxillo facial radiology最新文献

Objectives: Temporomandibular joint disorder (TMD) patients experience a variety of clinical symptoms, and magnetic resonance imaging (MRI) is the most effective tool for diagnosing temporomandibular joint (TMJ) disc displacement. This study aimed to develop a transformer-based deep learning model to generate T2-weighted (T2w) images from proton density-weighted (PDw) images, reducing MRI scan time for TMD patients.

Methods: A dataset of 7,226 images from 178 patients who underwent TMJ MRI examinations was used. The proposed model employed a generative adversarial network framework with a TransUNet architecture as the generator for image translation. Additionally, a disc segmentation decoder was integrated to improve image quality in the TMJ disc region. The model performance was evaluated using metrics such as the structural similarity index measure (SSIM), learned perceptual image patch similarity (LPIPS), and Fréchet inception distance (FID). Three experienced oral radiologists also performed a qualitative assessment through the mean opinion score (MOS).

Results: The model demonstrated high performance in generating T2w images from PDw images, achieving average SSIM, LPIPS, and FID values of 82.28%, 2.46, and 23.85, respectively, in the disc region. The model also obtained an average MOS score of 4.58, surpassing other models. Additionally, the model showed robust segmentation capabilities for the TMJ disc.

Conclusion: The proposed model using the transformer, complemented by an integrated disc segmentation task, demonstrated strong performance in MR image generation, both quantitatively and qualitatively. This suggests its potential clinical significance in reducing MRI scan times for TMD patients while maintaining high image quality.

Objectives: This study aimed to evaluate the accuracy of airway volume measurements obtained from cone-beam computed tomography (CBCT) images using various software programs, with a focus on assessing the performance of NemoStudio compared to other tools. The estimated volumes were compared with the volume of the solid model's cavity filled with water (gold standard).

Methods: A single 3D-printed airway model was created based on CBCT data and scanned ten times under identical conditions. Volume measurements were performed using semi-automatic segmentation in four software programs (NemoStudio, NNT Viewer, ITK-SNAP, and 3D Slicer). The results were compared to the gold standard using repeated measures ANOVA, Bland-Altman plots, and post hoc comparisons.

Results: Nemo Studio demonstrated a systematic bias and higher variability compared to the gold standard, resulting in lower accuracy than the other software programs. ITK-SNAP and 3D Slicer showed the highest agreement with the gold standard, while NNT Viewer also exhibited acceptable performance. Statistical analyses revealed significant differences in the accuracy of volume measurements among the software tools (P < 0.001). Bland-Altman plots highlighted Nemo Studio's broader limits of agreement, emphasizing its deviation from the gold standard.

Conclusion: Variability in airway volume measurement accuracy underscores the need for careful software selection and methodological standardization. Further refinement of segmentation algorithms is essential for improved consistency and reliability in clinical applications.

Advances in knowledge: This study provides the first evaluation of NemoStudio's volumetric accuracy for CBCT-based airway measurements, offering novel insights into software reliability and the impact of algorithm selection in clinical and academic settings.

Objectives: To evaluate patients' perceptions of the use of artificial intelligence (AI) in dental imaging diagnostics across six centers 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 center, 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-center differences were analyzed 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 centers, with participants from Ribeirão Preto being more likely to accept AI replacing dentists, whereas those from Aarhus and Tromsø expressed greater skepticism about AI's diagnostic capabilities. Higher levels of education and familiarity with AI were positively associated with more favorable views, provided that human supervision remained a key component.

Conclusions: Overall, patients favor 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 and facilitate the integration of AI into dental care.

Objective: To quantitatively and qualitatively compare directly two types of cisternography images for diagnosing trigeminal neuralgia (TN) using 3-T magnetic resonance imaging.

Methods: This prospective study recruited 64 patients with a clinical diagnosis or suspicion of TN. Patients were examined through the three-dimensional (3D) Constructive Interference in Steady State (CISS) and Sampling Perfection with Application-optimized Contrasts using different flip angle Evolutions (SPACE) sequences. Three radiologists quantitatively measured the signal intensity of the trigeminal nerve (cranial nerve V, CN5) (SICN5), cerebrospinal fluid (CSF) (SICSF), and contrast between CN5 and CSF (Cont.). Additionally, two radiologists qualitatively evaluated the basilar artery (BA), CN5, CSF, image artefacts, and overall image quality. Statistical analyses included paired-sample t-tests, non-parametric McNemar tests, and the Friedman test (significance set at p < 0.05).

Results: Mean SICN5 (p < 0.001), SICSF (p = 0.679), and Cont. (p < 0.001) were as follows: 203.08 ± 26.68, 936.03 ± 91, and 3.68 ± 0.74 in CISS; 46.80 ± 16.88, 940.61 ± 71.39, and 23.19 ± 14.52 in SPACE. Low-to-moderate CN5 and BA visibility was observed in all cases in CISS, while it was noted in one case for CN5 and in none for BA in SPACE (p < 0.001). Homogenous CSF and minor artefacts were observed in 14 cases in CISS, while it was seen in 52 cases for CN5 and 59 for BA in SPACE (p < 0.001). The overall image quality was scored as four in 57 cases in SPACE, while no cases received this score in CISS (p < 0.001).

Conclusions: SPACE provided better images than CISS for evaluating CN5 and prepontine cistern vascularity, indicating a valuable sequence for TN diagnosis.

Advances in knowledge: This study indicates that SPACE should be selected for TN diagnosis instead of CISS sequence.

Objectives: This meta-research assessed methodologies used for evaluating peri-implant marginal bone levels on digital periapical radiographs in randomized clinical trials published between 2019 and 2023.

Methods: Articles were searched in four databases. Data on methods for assessing peri-implant marginal bone levels were extracted. Risk of bias assessment was performed.

Results: During full-text reading, 108 out of 162 articles were excluded. Methodological issues accounted for these exclusions, including the absence of radiograph-type information, the lack of radiographic positioners, the missing anatomical references, and the use of panoramic radiographs or tomography. Fifty-four articles were included, most from Europe (70%) and university-based (74%). Radiographic positioners were specified in 54% of articles. Examiner calibration was unreported in 54%, with 69% lacking details. In 59%, no statistical measure assessed examiner agreement. Blinding was unreported or unused in 50%. Marginal bone level changes were the primary outcome of 61%. Most articles (59.3%) raised "some concerns" regarding bias, while 37% showed a high risk of bias, and only two articles (3.7%) demonstrated a low risk of bias.

Conclusions: Several limitations and areas for improvement were identified. Future studies should prioritize protocol registration, standardize radiographic acquisitions, specify examiner details, implement calibration and statistical measures for agreement, introduce blinding protocols, and maintain geometric calibration standards.

Objectives: Cysts in jaws may have similar radiographic features. However, it is important to clarify the diagnosis prior to surgery. The aim of this study was to compare the radiomic features of radicular cysts (RCs), dentigerous cysts (DCs), and odontogenic keratocysts (OKCs) as a non-invasive diagnostic alternative to biopsy.

Methods: In total, 161 odontogenic cysts diagnosed histopathologically (55 RCs, 53 DCs, and 53 OKCs) were included in the present study. Each cyst was semi-automatically segmented on CBCT images, and radiomic features were extracted by an observer. A second observer repeated 20% of the evaluations and the radiomic features. Those achieving an inter-observer agreement level above 0.850 were included in the study. Consequently, five shape-based and 22 textural features were investigated in the study. Statistical analysis was performed comparing both three cyst features and making pairwise comparisons.

Results: All features included in the study showed statistical differences between cysts, with the exception of one textural feature (NGTDM coarseness) (P < .05). However, only one shape-based feature (shericity) and one textural feature (GLSZM large area emphasis) were statistically different in pairwise comparisons of all three cysts (P < .05).

Conclusion: Radiomics features of the RCs, DCs, and OKCs showed significant differences, and may have the potential to be used as a non-invasive method in the differential diagnosis of cysts.

Objectives: Due to the increasing use of cone-beam CT (CBCT) in dentistry and considering the effects of radiation on radiosensitive organs, the aim of this study was to investigate the effect of shielding on absorbed dose of eyes, thyroid, and breasts in scans conducted with different parameters using 2 different fields of view (FOV).

Methods: Dose measurements were calculated on a tissue-equivalent female phantom by repeating each scanning parameter 3 times and placing at least 2 thermoluminescent dosimeters (TLD) on each organ, with the averages then taken. The same CBCT scans were performed in 2 different FOV with shielding including thyroid collar, radiation safety glasses, and lead apron and without shielding. The differences between them were analysed statistically. Descriptive statistics and the Wilcoxon test were used for data analysis.

Results: The difference between measurements with and without shielding was statistically significant for all scans (P < .001). The dose reduction associated with the use of shielding ranged from 26.81% to 52.95%. The dose related to the FOV has shown a significant increase, ranging from 8.30% to 623.54%, due to both the variation in the area affected by the primary beam on the organs and changes in the amount of radiation.

Conclusion: There are significant differences in the absorbed dose depending on shielding and FOV usage. Therefore, the CBCT imaging protocol should be optimized by the operator, and special attention should be paid to the use of thyroid collars and radiation safety glasses, considering their effects on image quality.

Objectives: The purpose of this study was to compare the image quality of ultra-high-resolution CT (U-HRCT) with that of conventional multidetector row CT (convCT) and demonstrate its usefulness in the dentomaxillofacial region.

Methods: Phantoms were helically scanned with U-HRCT and convCT scanners using clinical protocols. In U-HRCT, phantoms were scanned in super-high-resolution (SHR) mode, and hybrid iterative reconstruction (HIR) and filtered-back projection (FBP) techniques were performed using a bone kernel (FC81). The FBP technique was performed using the same kernel as in convCT (reference). Two observers independently evaluated the 54 resulting images using a 5-point scale (5 = excellent diagnostic image quality; 4 = above average; 3 = average; 2 = subdiagnostic; and 1 = unacceptable). The system performance function (SPF) was calculated for a comprehensive evaluation of the image quality using the task transfer function and noise power spectrum. Statistical analysis using the Kruskal-Wallis test was performed to compare the image quality among the 3 protocols.

Results: The observers assigned higher scores to images acquired with the SHRHIR and SHRFBP protocols than to those acquired with the reference (P < 0.0001 and P < 0.0001, respectively). The relative SPF value at 1.0 cycles/mm in SHRHIR and SHRFBP compared to the reference protocol were 151.5% and 45.6%, respectively.

Conclusions: Through phantom experiments, this study demonstrated that U-HRCT can provide superior-quality images compared to conventional CT in the dentomaxillofacial region. The development of a better image reconstruction method is required to improve image quality and optimize the radiation dose.

The aim of this technical report was to assess whether the "Radiological Report" tool within the Artificial Intelligence (AI) software Diagnocat can achieve a satisfactory level of performance comparable to that of experienced dentomaxillofacial radiologists in interpreting cone-beam CT scans. Ten cone-beam CT scans were carefully selected and analysed using the AI tool, and they were also evaluated by two dentomaxillofacial radiologists. Observations related to tooth numeration, alterations in dental crowns, roots, and periodontal tissues were documented and subsequently compared to the AI findings. Kappa statistics, along with their corresponding 95% confidence intervals, were calculated to ascertain the degree of agreement. The agreement between the AI tool and the radiologists ranged from substantial to nearly perfect for identifying teeth, determining the number of roots and canals, assessing crown conditions, and detecting endodontic treatments. However, for tasks such as classifying bone loss, identifying posts, evaluating the quality of fillings, and appraising the situation of periodontal spaces, the agreement was deemed slight. In conclusion, the "radiological report" tool of the Diagnocat demonstrates satisfactory performance in reliably identifying teeth, roots, canals, assessing crown conditions, and detecting endodontic treatment. However, further investigations are needed to evaluate the tool's effectiveness in diagnosing posts, assessing the condition and quality of fillings, and determining the status of periodontal spaces.

Objective: To develop an accurate method for converting dose-area product (DAP) to patient dose for dental cone-beam computed tomography (CBCT) using deep learning.

Methods: A total of 24 384 CBCT exposures of an adult phantom were simulated with PCXMC 2.0, using permutations of tube voltage, filtration, source-isocenter distance, beam width/height, and isocenter position. Equivalent organ doses as well as DAP values were recorded. Next, using the aforementioned scan parameters as inputs, neural networks (NN) were trained using Keras for estimating the equivalent dose per DAP for each organ. Two methods were explored for positional input features: (1) "Coordinate" mode, which uses the (continuous) XYZ coordinates of the isocentre, and (2) "AP/JAW" mode, which uses the (categorical) anteroposterior and craniocaudal position. Each network was trained, validated, and tested using a 3/1/1 data split. Effective dose (ED) was calculated from the combination of NN outputs using ICRP 103 tissue weighting factors. The performance of the resulting NN models for estimating ED/DAP was compared with that of a multiple linear regression (MLR) model as well as direct conversion coefficients (CC).

Results: The mean absolute error (MAE) for organ dose/DAP on the test data ranged from 0.18% (bone surface) to 2.90% (oesophagus) in "Coordinate" mode and from 2.74% (red bone marrow) to 14.13% (brain) in "AP/JAW" mode. The MAE for ED was 0.23% and 4.30%, respectively, for the two modes, vs. 5.70% for the MLR model and 20.19%-32.67% for the CCs.

Conclusions: NNs allow for an accurate estimation of patient dose based on DAP in dental CBCT.