Application of deep learning and feature selection technique on external root resorption identification on CBCT images.

BACKGROUND: Artificial intelligence has been proven to improve the identification of various maxillofacial lesions. The aim of the current study is two-fold: to assess the performance of four deep learning models (DLM) in external root resorption (ERR) identification and to assess the effect of combining feature selection technique (FST) with DLM on their ability in ERR identification. METHODS: External root resorption was simulated on 88 extracted premolar teeth using tungsten bur in different depths (0.5 mm, 1 mm, and 2 mm). All teeth were scanned using a Cone beam CT (Carestream Dental, Atlanta, GA). Afterward, a training (70%), validation (10%), and test (20%) dataset were established. The performance of four DLMs including Random Forest (RF) + Visual Geometry Group 16 (VGG), RF + EfficienNetB4 (EFNET), Support Vector Machine (SVM) + VGG, and SVM + EFNET) and four hybrid models (DLM + FST: (i) FS + RF + VGG, (ii) FS + RF + EFNET, (iii) FS + SVM + VGG and (iv) FS + SVM + EFNET) was compared. Five performance parameters were assessed: classification accuracy, F1-score, precision, specificity, and error rate. FST algorithms (Boruta and Recursive Feature Selection) were combined with the DLMs to assess their performance. RESULTS: RF + VGG exhibited the highest performance in identifying ERR, followed by the other tested models. Similarly, FST combined with RF + VGG outperformed other models with classification accuracy, F1-score, precision, and specificity of 81.9%, weighted accuracy of 83%, and area under the curve (AUC) of 96%. Kruskal Wallis test revealed a significant difference (p = 0.008) in the prediction accuracy among the eight DLMs. CONCLUSION: In general, all DLMs have similar performance on ERR identification. However, the performance can be improved by combining FST with DLMs.

Characterization of the root and canal anatomy of maxillary premolar teeth in an Iraqi subpopulation: a cone beam computed tomography study.

This study aims to evaluate the number of roots and root canal morphology types of maxillary premolars in relation to a patient's gender and age in an Iraqi population using two classification systems. Cone beam computed tomography (CBCT) scans of 1116 maxillary premolars from 385 patients were evaluated for the number of roots and root canal morphology types according to Vertucci's classification and Ahmed et al. classification systems. Differences in the number of roots and root canal morphology types with regard to tooth type, patients' gender and age groups were evaluated and the degree of bilateral symmetry was determined. Chi-squared test was used for statistical analysis. About 51.1% of the 1st premolars were double rooted. The majority (87.9%) of the 2nd premolars were single rooted. The three-rooted form presented in only 1.2% and 0.7% of the 1st and 2nd premolars, respectively. Vertucci Type IV (Ahmed et al. code 2MaxP B1P1) and Vertucci Type I (Ahmed et al. code 1MaxP1) were the most common canal morphology types in the 1st and 2nd premolars, respectively. Females showed a lower number of roots and a higher prevalence of Vertucci Type I configuration (P < 0.05). Younger age groups showed a higher prevalence of Vertucci Type I configuration (P < 0.05). Bilateral symmetry was seen in more than half of the maxillary premolars. There is a considerable variation in the number of roots and root canal configurations of maxillary premolars in the studied Iraqi population, with a significant difference by gender and age groups. Ahmed et al. classification provided more accurate presentation of the root and canal anatomy in maxillary premolars compared to Vertucci's classification.

Efficacy and bone-contact biocompatibility of glass ionomer cement as a biomaterial for bone regeneration: A systematic review.

Bone regeneration is a rapidly growing field that seeks to develop new biomaterials to regenerate bone defects. Conventional bone graft materials have limitations, such as limited availability, complication, and rejection. Glass ionomer cement (GIC) is a biomaterial with the potential for bone regeneration due to its bone-contact biocompatibility, ease of use, and cost-effectiveness. GIC is a two-component material that adheres to the bone and releases ions that promote bone growth and mineralization. A systematic literature search was conducted using PubMed-MEDLINE, Scopus, and Web of Science databases and registered in the PROSPERO database to determine the evidence regarding the efficacy and bone-contact biocompatibility of GIC as bone cement. Out of 3715 initial results, thirteen studies were included in the qualitative synthesis. Two tools were employed in evaluating the Risk of Bias (RoB): the QUIN tool for assessing in vitro studies and SYRCLE for in vivo. The results indicate that GIC has demonstrated the ability to adhere to bone and promote bone growth. Establishing a chemical bond occurs at the interface between the GIC and the mineral phase of bone. This interaction allows the GIC to exhibit osteoconductive properties and promote the growth of bone tissue. GIC's bone-contact biocompatibility, ease of preparation, and cost-effectiveness make it a promising alternative to conventional bone grafts. However, further research is required to fully evaluate the potential application of GIC in bone regeneration. The findings hold implications for advancing material development in identifying the optimal composition and fabrication of GIC as a bone repair material.

Impact of early biofilm on the assessment of initial enamel erosion with swept-source optical coherence tomography.

Objective: This study examined the impact of early biofilm on the tooth surface, during the assessment of initial enamel erosion using swept-source optical coherence tomography (SS-OCT). Method and materials: Forty-five enamel windows of 2 × 4 mm2 were prepared on 23 extracted human teeth. The specimens were exposed to citric acid (pH 3.2) for 30 minutes and randomly divided into three groups (n = 15): Group 1, no biofilm; Group 2, 1-day-old biofilm; and Group 3, 3-day-old biofilm. Specimens in Groups 2 and 3 were inoculated with oral bacteria (Streptococcus sanguinis, Streptococcus mitis, and Actinomyces naeslundii) to produce early laboratory-cultivated biofilms for 1 and 3 days respectively. Surface microhardness (SMH) measurements were taken at pre- (t1) and post-erosion (t2); and SS-OCT scans were done at t1, t2, and post-biofilm cultivation (t3). Integrated reflectivity (IR) of the tooth-air interface (IRsurface) and enamel (IRenamel) were computed from the mean A-scans. Statistical analysis was performed using paired t tests and one-way ANOVA (α = .05). Results: A significant increase in IRenamel was observed at t2 (P < .05). At t3, IRsurface between Group 1 (control) and Group 2 (P = .012) as well as Group 3 (P = .001) were significantly different. Significant variances in IRenamel were perceived between t2 and t3 for Groups 2 and 3 but not for Group 1. Conclusion: As early biofilm affected SS-OCT assessment of initial enamel erosion, they should be removed from the tooth surface prior to OCT procedures..