The influence of different Cheek and lip retractors and emissivity on intraoral infrared thermography.

OBJECTIVES: To assess the influence of two conventional and one adapted cheek and lip retractors and three emissivity setting values on intraoral infrared thermography (IT) temperature values. METHODS: The sample was composed by 50 volunteers. Three cheek and lip retractors were tested: Group 1 - flex retractor (FR); Group 2 - FR adapted with Styrofoam; Group 3 - U-type retractor (UR) for cheek and lip. All thermograms were acquired using FLIR T650 infrared camera. A set of three thermograms in frontal norm were acquired for each lip and cheek retractor at 0.91, 0.96 and 0.98Ɛ, with an interval of 15 minutes between each set of images to avoid thermal interference. All images were assessed by two observers. The ROIs' mean temperature of the four upper incisors were recorded. Two-way ANOVA and Sidak posttest were used for data assessment with a significance level of 5%. RESULTS: Group 3 showed higher mean temperature than groups 1 and 2 at all emissivity settings for all assessed teeth (p < 0.05). 0.91Ɛ showed higher temperature than 0.96Ɛ and 0.98Ɛ for all assessed variables (p < 0.01). Contralateral teeth assessed using Group 3 at 0.91Ɛ showed statistical differences between each other (p < 0.05). No statistical difference was observed between contralateral teeth assessed using Group 1 and 2 at 0.96 and 0.98Ɛ (p > 0.05). CONCLUSIONS: The choice of cheek and lip retractor and emissivity setting can interfere on intraoral IT temperature values. U-type cheek and lip retractor and 0.91Ɛ setting should not be used for IT image acquisition when assessing dental tissues.

Artificial intelligence and infrared thermography as auxiliary tools in the diagnosis of temporomandibular disorder.

OBJECTIVE: To assess three machine learning (ML) attribute extraction methods: radiomic, semantic and radiomic-semantic association on temporomandibular disorder (TMD) detection using infrared thermography (IT); and to determine which ML classifier, KNN, SVM and MLP, is the most efficient for this purpose. METHODS AND MATERIALS: 78 patients were selected by applying the Fonseca questionnaire and RDC/TMD to categorize control patients (37) and TMD patients (41). IT lateral projections of each patient were acquired. The masseter and temporal muscles were selected as regions of interest (ROI) for attribute extraction. Three methods of extracting attributes were assessed: radiomic, semantic and radiomic-semantic association. For radiomic attribute extraction, 20 texture attributes were assessed using co-occurrence matrix in a standardized angulation of 0°. The semantic features were the ROI mean temperature and pain intensity data. For radiomic-semantic association, a single dataset composed of 28 features was assessed. The classification algorithms assessed were KNN, SVM and MLP. Hopkins's statistic, Shapiro-Wilk, ANOVA and Tukey tests were used to assess data. The significance level was set at 5% (p < 0.05). RESULTS: Training and testing accuracy values differed statistically for the radiomic-semantic association (p = 0.003). MLP differed from the other classifiers for the radiomic-semantic association (p = 0.004). Accuracy, precision and sensitivity values of semantic and radiomic-semantic association differed statistically from radiomic features (p = 0.008, p = 0.016 and p = 0.013). CONCLUSION: Semantic and radiomic-semantic-associated ML feature extraction methods and MLP classifier should be chosen for TMD detection using IT images and pain scale data. IT associated with ML presents promising results for TMD detection.

Assessment of the Influence of Different Intracanal Materials on the Detection of Root Fracture in Birooted Teeth by Cone-beam Computed Tomography.

INTRODUCTION: The aim of this study was to assess 2 cone-beam computed tomographic systems on the detection of artificially induced vertical root fractures (VRFs) and artifact intensity using birooted teeth restored with different intracanal materials. METHODS: The sample consisted of 20 extracted birooted premolars. Root fracture was induced in half of the sample. Seven intracanal material combinations were used in each tooth, 1 at a time: unrestored, gutta-percha, a buccal root with gutta-percha and a lingual root with a fiberglass post, a buccal root with gutta-percha and a lingual root with a metal core fiberglass post, fiberglass posts, metal core fiberglass posts, and NiCr posts. Cone-beam computed tomographic scans were acquired using CS 9000 3D (Carestream Dental Rochester, NY) and OP300 (Instrumentarium Dental Inc, Tuusula, Finland) units. Exposure parameters were fixed at 90 kV and 8 mA. The voxel size and field of view were set at 0.085 mm and 5 × 5 cm for OP300 and 0.076 mm and 5 × 3.75 cm for CS 9000, respectively. Two observers assessed all images using a 5-point confidence scale for VRF detection and a 4-point score for artifact interference. The sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve were compared using 2-way analysis of variance and the Tukey test (α = 0.05). Artifact interference was evaluated by descriptive statistics and the chi-square test. RESULTS: There were significant differences between scanners (P > .05) and among the different intracanal material groups (OP300) (P < .05) for specificity. When a metal post was present in both roots, severe artifact interference was observed in all images. CONCLUSIONS: CS 9000 3D presented better performance than OP300 on VRF detection of endodontically treated teeth. Unrestored teeth and teeth filled with fiberglass posts were considered the groups with the lowest artifact interference and the highest VRF detection results.

Assessment of Cone-beam Computed Tomographic Artifacts from Different Intracanal Materials on Birooted Teeth.

INTRODUCTION: This study quantitatively evaluated the type and amount of image artifacts generated by different intracanal materials in birooted teeth scanned at different exposure parameters. METHODS: The sample consisted of 15 birooted premolars. Seven different intracanal material combinations were used in each tooth one at a time: (1) roots without intracanal materials, (2) roots with gutta-percha, (3) a buccal root with gutta-percha and a lingual root with a fiberglass post, (4) a buccal root with gutta-percha and a lingual root with a metal core fiberglass post, (5) buccal and lingual roots with fiberglass posts, (6) buccal and lingual roots with metal core fiberglass posts, and (7) buccal and lingual roots with NiCr metal posts. Cone-beam computed tomographic scans were acquired using a CS 9000 unit (Carestream Dental, Atlanta, GA). An image of each tooth was captured under 5 exposure parameters: 2.5, 4, 6.3, 8, and 12 mA. The voxel size, field of view, and tube voltage were fixed at 0.076 mm, 5 × 3.75 cm, and 75 kV. We assessed each artifact quantitatively using ImageJ's threshold tool (National Institutes of Health, Bethesda, MD) to determine the hypodense and hyperdense artifact areas within 8-bit images extracted from the scans. All analyses were conducted with a 95% confidence level (α <0.05). RESULTS: The inferential analysis showed that roots filled with metal posts presented the highest amount of hypodense and hyperdense artifacts, whereas fiberglass post in both roots presented fewer artifacts. All materials presented more hypodense than hyperdense artifact formation. Overall, the low-exposure settings presented fewer artifacts and higher values of preserved dental images. CONCLUSIONS: Low-exposure protocols and fiberglass posts presented fewer image artifacts in CBCT scans.