A Deep Learning Approach to Segment and Classify C-Shaped Canal Morphologies in Mandibular Second Molars Using Cone-beam Computed Tomography.

INTRODUCTION: The identification of C-shaped root canal anatomy on radiographic images affects clinical decision making and treatment. The aims of this study were to develop a deep learning (DL) model to classify C-shaped canal anatomy in mandibular second molars from cone-beam computed tomographic (CBCT) volumes and to compare the performance of 3 different architectures. METHODS: U-Net, residual U-Net, and Xception U-Net architectures were used for image segmentation and classification of C-shaped anatomies. Model training and validation were performed on 100 of a total of 135 available limited field of view CBCT images containing mandibular molars with C-shaped anatomy. Thirty-five CBCT images were used for testing. Voxel-matching accuracy of the automated labeling of the C-shaped anatomy was assessed with the Dice index. The mean sensitivity of predicting the correct C-shape subcategory was calculated based on detection accuracy. One-way analysis of variance and post hoc Tukey honestly significant difference tests were used for statistical evaluation. RESULTS: The mean Dice coefficients were 0.768 ± 0.0349 for Xception U-Net, 0.736 ± 0.0297 for residual U-Net, and 0.660 ± 0.0354 for U-Net on the test data set. The performance of the 3 models was significantly different overall (analysis of variance, P = .000779). Both Xception U-Net (Q = 7.23, P = .00070) and residual U-Net (Q = 5.09, P = .00951) performed significantly better than U-Net (post hoc Tukey honestly significant difference test). The mean sensitivity values were 0.786 ± 0.0378 for Xception U-Net, 0.746 ± 0.0391 for residual U-Net, and 0.720 ± 0.0495 for U-Net. The mean positive predictive values were 77.6% ± 0.1998% for U-Net, 78.2% ± 0.0.1971% for residual U-Net, and 80.0% ± 0.1098% for Xception U-Net. The addition of contrast-limited adaptive histogram equalization had improved overall architecture efficacy by a mean of 4.6% (P < .0001). CONCLUSIONS: DL may aid in the detection and classification of C-shaped canal anatomy.

Regenerative Capacity of Dental Pulp Stem Cells: A Systematic Review.

OBJECTIVES: The dental pulp contains undifferentiated mesenchymal cells, blood vessels and so on, which are responsible for routine functions of a tooth. The determination of stemness and regenerative properties using biomarkers and further application in routine practice may unravel its potential. MATERIALS AND METHODS: Inclusion criteria-original research articles published in English, from 2000 to 2019, were collected both manually and by electronic search from databases of Cochrane, Medline, Embase, and PubMed. Exclusion criteria-articles other than English and review manuscripts were omitted. The shortlisted articles were reviewed for specific biomarkers, to assess the regenerative potential, stemness, and lineage of dental pulp stem cells. RESULTS: Of 512 articles, 64 were selected and reviewed to determine the mesenchymal, neurogenic, vasculogenic, hematopoietic, and stem cell potential. On the basis of the search analysis, a panel of markers was proposed. CONCLUSION: The application of proposed markers, on a pulpectomized tissue derived from human teeth, may be helpful to determine the regenerative potential and the usefulness in regenerative medicine and tissue engineering.

Influence of Tooth Factors and Procedural Errors on the Incidence and Severity of Post-Endodontic Pain: A Prospective Clinical Study.

The objective of this prospective study was to assess tooth-related factors that play a role in the incidence of postoperative pain (PP) and determining if procedural errors influence PP occurrence. A total of 442 patients referred for root canal treatment met the inclusion criteria and were included in this prospective study. The same protocol was used in all root canal treatments. Patient, tooth, treatment related factors and the occurrence of procedural errors were registered. Incidence and intensity of PP was assessed at 24 and 48 h by telephonic interview and in person seven and 15 days after treatment. A logistic and ordinal regression analysis was used to assess the role of patient, tooth and treatment related factors in the incidence and intensity of PP, respectively. Preoperative and intraoperative factors differently affected the incidence of PP at the different time intervals. The presence of procedural errors did not significantly influence PP occurrence. The presence of preoperative pain and the need of additional anesthesia during treatment were associated with higher incidence of PP 24 and 48 h after treatment; the extent of apical enlargement played a significant role in the presence of PP after seven days of treatment; and the excessive occlusal load induced by the absence of a contralateral tooth was the only factor related to the maintenance of PP up to 15 days. In conclusion, the presence of preoperative pain, the need of additional anesthesia during treatment, the extent of apical enlargement and the excessive occlusal load induced by the absence of a contralateral tooth were related to a higher incidence of PP.

A subset of epithelioid and spindle cell rhabdomyosarcomas is associated with TFCP2 fusions and common ALK upregulation.

Rhabdomyosarcomas with TFCP2 fusions represent an emerging subtype of tumors, initially discovered by RNA-sequencing. We report herein the clinicopathological, transcriptional, and genomic features of a series of 14 cases. Cases were retrospectively and prospectively recruited and studied by immunohistochemistry (MYF4, MYOD1, S100, AE1/E3, ALK), fluorescence in situ hybridization with TFCP2 break-apart probe (n = 10/14), array-comparative genomic hybridization (Agilent), whole RNA-sequencing (Truseq Exome, Illumina), or anchored multiplex PCR-based targeted next-generation sequencing (Archer® FusionPlex® Sarcoma kit). Patient's age ranged between 11 and 86 years, including 5 pediatric cases. Tumors were located in the bone (n = 12/14) and soft tissue (n = 2/14). Most bone tumors invaded surrounding soft tissue. Craniofacial bones were over-represented (n = 8/12). Median survival was 8 months and five patients are currently alive with a median follow-up of 20 months. Most tumors displayed a mixed spindle cell and epithelioid pattern with frequent vesicular nuclei. All tumors expressed keratins and showed a rhabdomyogenic phenotype (defined as expression of MYF4 and/or MYOD1). ALK was overexpressed in all but three cases without underlying ALK fusion on break-apart FISH (n = 5) nor next-generation sequencing (n = 14). ALK upregulation was frequently associated with an internal deletion at genomic level. TFCP2 was fused in 5' either to EWSR1 (n = 6) or FUS (n = 8). EWSR1 was involved in both soft tissue cases. FISH with TFCP2 break-apart probe was positive in all tested cases (n = 8), including one case with unbalanced signal. On array-CGH, all tested tumors displayed complex genetic profiles with genomic indexes ranging from 13 to 107.55 and recurrent CDKN2A deletions. FET-TFCP2 rhabdomyosarcomas clustered together and distinctly from other rhabdomyosarcomas subgroups. Altogether, our data confirm and expand the spectrum of the new family of FET-TFCP2 rhabdomyosarcomas, which are associated with a predilection for the craniofacial bones, an aggressive course, and recurrent pathological features. Their association with ALK overexpression might represent a therapeutic vulnerability.