External validation of cone-beam computed tomography- and panoramic radiography-featured prediction models for inferior alveolar nerve injury after lower third molar removal: proposal of a risk calculator.

We previously developed basic and extended models to predict inferior alveolar nerve injuries (IANI) after lower third molar (LM3) removal based on cone-beam computed tomography (CBCT) images. Although these models comprised predictors, including increased age and inferior alveolar canal-related CBCT factors, external validations were lacking. Therefore, this study externally validated these models and compared them with other related models based on their performance. Original and newly validated samples included patients who underwent LM3 removal following CBCT. Subsequently, 39 and 25 patients with IANI, then 457 and 295 randomly selected patients without IANI were chosen of the observed 1573 and 1052 patients, respectively. CBCT- and panoramic radiograph (PAN)-featured models were validated. Then, models' discrimination and calibration abilities were assessed using C-statistics and calibration plots, respectively. Brier scores were also quantified, after which logistic recalibration was achieved to optimize calibration, and a risk calculator was developed. During the external validation, the extended model exhibited the best C-statistic (0.822) and Brier score (0.064), whereas two CBCT- and two PAN-featured models showed lower performances with C-statistics (0.764, 0.706, 0.584, and 0.627) and Brier scores (0.069, 0.074, 0.075, and 0.072). Besides, all models showed a tendency to overpredict its high-risk range. However, recalibration of the extended model resulted in excellent calibration performance. CBCT-featured models, especially the extended model, conclusively showed a superior predictive performance to PAN models. Therefore, the risk calculator on the extended CBCT model is proposed to be a clinical decision-aid tool that preoperatively predicts IANI risk.

Deep learning model for the automated evaluation of contact between the lower third molar and inferior alveolar nerve on panoramic radiography.

Background/purpose: In lower third molar (LM3) surgery, panoramic radiography (PAN) is important for the initial assessment of the anatomical association between LM3 and the inferior alveolar nerve (IAN). This study aimed to develop a deep learning model for the automated evaluation of the LM3-IAN association on PAN. Further, its performance was compared with that of oral surgeons using original and external datasets. Materials and methods: In total, 579 panoramic images of LM3 from 384 patients in the original dataset were utilized. The images were divided into 483 images for the training dataset and 96 for the testing dataset at a ratio of 83:17. The external dataset comprising 58 images from an independent institution was used for testing only. The LM3-IAN associations on PAN were categorized into direct or indirect contact based on cone-beam computed tomography (CBCT). The You Only Look Once (YOLO) version 3 algorithm, a fast object detection system, was applied. To increase the amount of training data for deep learning, PAN images were augmented using the rotation and flip techniques. Results: The final YOLO model had high accuracy (0.894 in the original dataset and 0.927 in the external dataset), recall (0.925, 0.919), precision (0.891, 0.971), and f1-score (0.908, 0.944). Meanwhile, oral surgeons had lower accuracy (0.628, 0.615), recall (0.821, 0.497), precision (0.607, 0.876), and f1-score (0.698, 0.634). Conclusion: The YOLO-driven deep learning model can help oral surgeons in the decision-making process of applying additional CBCT to confirm the LM3-IAN association based on PAN images.

Mutations of the APC, beta-catenin, and axin 1 genes and cytoplasmic accumulation of beta-catenin in oral squamous cell carcinoma.

PURPOSE: The Wnt pathway is involved in carcinogenesis and three regulatory genes of the Wnt pathway, APC, beta-catenin and Axin are mutated in some primary human cancers. Mutations in these genes can impair the down regulation of beta-catenin, which results in the stabilization of beta-catenin, accumulation of free beta-catenin and subsequent activation of the Wnt pathway. To clarify the genetic alterations of components of the Wnt pathway in oral squamous cell carcinoma (SCC), we examined mutations in the APC, beta-catenin and Axin genes and subcellular localization of beta-catenin. METHODS: 20 oral SCC tissues and four cell lines derived from oral SCC were used. Mutational analysis was performed by a single-strand conformation polymorphism (SSCP) method and direct sequencing analysis. The samples were also examined by immunohistochemical staining and immunoblot analysis. RESULTS: In 3 of 4 cell lines, mutations were observed in the APC and Axin1 genes without amino acid substitutions. In a clinical sample, a mutation in the Axin1 gene was detected; a T insertion at codon 250 resulted in the formation of a stop codon at codon 259. In addition, cytoplasmic accumulation of beta-catenin was observed in 3 (75%) of 4 cell lines and 18 (90%) of 20 cancer tissue samples. CONCLUSION: The Axin1 gene may be one of the mutational target in oral SCC. In addition, the cytoplasmic accumulation of beta-catenin is a common characteristic of oral SCC, but is not closely associated with mutational alterations in the APC, beta-catenin and Axin1 genes.

Induction of instability of p34(cdc2) expression by treatment with cisplatin (CDDP) in mouse teratocarcinoma F9 cells.

p34(cdc2) is a protein kinase that plays an important role in the control of the cell cycle and regulation of activity of the tumor suppressor gene. Previously, we demonstrated that cisplatin (CDDP) induced growth suppression resulting in differentiation of teratocarcinoma F9 cells. In the present study, we investigated the mechanism of cell cycle retardation by CDDP in F9 cells, focusing on p34(cdc2). After the induction of differentiation with CDDP, F9 cells were arrested at the late S+G2/M. After treatment with CDDP, the level of the expression of cdc2 mRNA did not change. However, the half-life of p34(cdc2) was greatly reduced, thus, the level of p34(cdc2) protein was decreased. These findings suggest that the cell cycle arrest by CDDP is due partly to the induced instability of p34(cdc2) in F9 cells.