Cutting guides in mandibular tumor ablation: Are we as accurate as we think?

Purpose: Tumor margin status is critical in local tumor recurrence and is a significant prognostic factor in head and neck cancer survival. With the introduction of computer-assisted surgical planning, one of the main challenges is the accurate positioning of the surgical cutting guide but there is limited evidence of the accuracy of the 3D cutting guides in mimicking virtually planned osteotomy. This study evaluates the accuracy of osteotomy lines produced by 3D-printed cutting guides and assesses the overall accuracy of mandibular reconstruction. Material and Methods: The pre and postoperative 3D models were aligned using an automated surface registration feature based on the iterative closest point algorithm. The differences in osteotomy line deviation, linear and angle measurements, and 3D volume quantification of the pre and post models were measured. Results: We included 14 patients (8 men and 6 women with ages ranging from 13 to 75 years) with a segmental mandibular resection who met all of the inclusion criteria. The smallest defect size was 4.4 cm, the largest defect was 12.2 cm, and the average was 7.30 cm +/- 2.80 cm. The average deviation between virtually planned osteotomy and actual surgical osteotomy was 1.52 +/-1.02 mm. No covariates were associated with increased inaccuracy of the 3D-printed cutting guides. Conclusion: The finding of this study suggests that virtual surgical planning is an unambiguous paradigm shift in the predictability of the surgical plan and achievement of the reconstruction goals. The 3D-printed cutting guides are a very accurate and reliable tool in translating virtual ablation plans to an actual surgical resection margin.

AggregateNet: A deep learning model for automated classification of cervical vertebrae maturation stages.

OBJECTIVE: A study of supervised automated classification of the cervical vertebrae maturation (CVM) stages using deep learning (DL) network is presented. A parallel structured deep convolutional neural network (CNN) with a pre-processing layer that takes X-ray images and the age as the input is proposed. METHODS: A total of 1018 cephalometric radiographs were labelled and classified according to the CVM stages. The images were separated according to gender for better model-fitting. The images were cropped to extract the cervical vertebrae automatically using an object detector. The resulting images and the age inputs were used to train the proposed DL model: AggregateNet with a set of tunable directional edge enhancers. After the features of the images were extracted, the age input was concatenated to the output feature vector. To have the parallel network not overfit, data augmentation was used. The performance of our CNN model was compared with other DL models, ResNet20, Xception, MobileNetV2 and custom-designed CNN model with the directional filters. RESULTS: The proposed innovative model that uses a parallel structured network preceded with a pre-processing layer of edge enhancement filters achieved a validation accuracy of 82.35% in CVM stage classification on female subjects, 75.0% in CVM stage classification on male subjects, exceeding the accuracy achieved with the other DL models investigated. The effectiveness of the directional filters is reflected in the improved performance attained in the results. If AggregateNet is used without directional filters, the test accuracy decreases to 80.0% on female subjects and to 74.03% on male subjects. CONCLUSION: AggregateNet together with the tunable directional edge filters is observed to produce higher accuracy than the other models that we investigated in the fully automated determination of the CVM stages.

Fully automated determination of the cervical vertebrae maturation stages using deep learning with directional filters.

INTRODUCTION: We aim to apply deep learning to achieve fully automated detection and classification of the Cervical Vertebrae Maturation (CVM) stages. We propose an innovative custom-designed deep Convolutional Neural Network (CNN) with a built-in set of novel directional filters that highlight the edges of the Cervical Vertebrae in X-ray images. METHODS: A total of 1018 Cephalometric radiographs were labeled and classified according to the Cervical Vertebrae Maturation (CVM) stages. The images were cropped to extract the cervical vertebrae using an Aggregate Channel Features (ACF) object detector. The resulting images were used to train four different Deep Learning (DL) models: our proposed CNN, MobileNetV2, ResNet101, and Xception, together with a set of tunable directional edge enhancers. When using MobileNetV2, ResNet101 and Xception, data augmentation is adopted to allow adequate network complexity while avoiding overfitting. The performance of our CNN model was compared with that of MobileNetV2, ResNet101 and Xception with and without the use of directional filters. For validation and performance assessment, k-fold cross-validation, ROC curves, and p-values were used. RESULTS: The proposed innovative model that uses a CNN preceded with a layer of tunable directional filters achieved a validation accuracy of 84.63%84.63% in CVM stage classification into five classes, exceeding the accuracy achieved with the other DL models investigated. MobileNetV2, ResNet101 and Xception used with directional filters attained accuracies of 78.54%, 74.10%, and 80.86%, respectively. The custom-designed CNN method also achieves 75.11% in six-class CVM stage classification. The effectiveness of the directional filters is reflected in the improved performance attained in the results. If the custom-designed CNN is used without the directional filters, the test accuracy decreases to 80.75%. In the Xception model without the directional filters, the testing accuracy drops slightly to 79.42% in the five-class CVM stage classification. CONCLUSION: The proposed model of a custom-designed CNN together with the tunable Directional Filters (CNNDF) is observed to provide higher accuracy than the commonly used pre-trained network models that we investigated in the fully automated determination of the CVM stages.

The Quality of the Evidence in Craniomaxillofacial Trauma: Are We Making Progress?

PURPOSE: The purpose of this study is to evaluate the level of evidence in the craniomaxillofacial trauma literature. The secondary purpose is to determine if improvement in the quality of evidence has been made over the past 2 decades. MATERIALS AND METHODS: This retrospective cohort study analyzed clinical articles published in select craniomaxillofacial surgery journals. Patient-focused articles were selected. The year of publication (1999, 2009, 2019) was the primary predictor variable. Secondary predictor variables included study funding status, anatomical facial region (eg maxilla, mandible, and so on), specific journal, and journal impact factor. The level of evidence was determined using the Center of Evidence-Based Medicine criteria, which served as the outcome variable. The statistical analysis included descriptive, bivariate, and regression analysis. RESULTS: The sample consisted of 469 craniomaxillofacial trauma articles, with 13.2% being published in 1999, 44.1% in 2009, and 42.6% in 2019. The majority of the studies (77.5%) were published in 4 journals. The impact factor varied among the journals with a significant improvement in the mean impact factor from (0.89 ± 0.29) in 1999 to (1.4 ± 0.47) in 2009 and a slight decline in 2019 (1.26 ± 0.71). Mandibular fractures (31.6%) and orbital trauma (26%) were the most investigated topics. Level 4 studies accounted for 67.4% of the sample, with level 3 evidence of 4.7%, level 2 of 22.6%, and level 1 of 5.3% of the included studies. Significant progress in the level of evidence has been made from 1999 but not since that time (P = .002). It is unclear why this may be but sheds light on the need for further development of high quality studies. Finally, a higher quality of evidence is associated with journal impact factor (odds ratio  = 1.9; P < .01) and funded research (odds ratio = 4.69; P = .02). CONCLUSIONS: While there has been some improvement in the level of evidence in the craniomaxillofacial trauma literature over the past 2 decades, the current quality remains low, and further progress is needed to support the practice of evidence-based medicine.

Nasotracheal Intubation: The Preferred Airway in Oral Cavity Microvascular Reconstructive Surgery?

PURPOSE: We aimed to describe the safety and effectiveness of nasotracheal intubation (NTI) in a cohort of patients undergoing reconstruction of oral cavity defects with free tissue transfer (FTT). MATERIALS AND METHODS: We implemented a retrospective cohort study and enrolled a sample composed of consecutive patients undergoing FTT reconstruction of oral cavity, maxillary, or mandibular defects between 2013 and 2017. These patients were all subject to a newly developed enhanced recovery-after-surgery protocol. The primary outcome measurement was hospital length of stay (LOS). The secondary outcome variables were the duration of mechanical ventilation, intensive care unit (ICU) LOS, need for gastrostomy, and airway-related complications directly associated with either NTI or tracheostomy. Descriptive statistics and a multivariate logistic regression analysis were completed. RESULTS: The sample was composed of 141 patients who had undergone oral cavity FTT for both benign and malignant diseases (NTI, n = 111; tracheostomy, n = 30). Patients managed with NTI had a statistically significantly shorter hospital LOS (8 days vs 15.5 days, P < .0001) and ICU LOS (1 day vs 2 days, P = .0006), as well as a decreased requirement for gastrostomy (17.1% vs 76.7%, P < .0001). Airway-related complications were rare in both the tracheostomy (13.3%) and NTI (3.6%) groups. Multivariate analysis showed that patients undergoing tracheostomy were 3.14 (P = .004) times more likely to have a prolonged hospitalization and 10.4 (P < .0001) times more likely to require a gastrostomy. A sensitivity analysis of only patients with malignant diagnoses had similar statistically significant results. The delayed tracheostomy rate in the NTI group was 3.6%. CONCLUSIONS: To date, this is the largest study to evaluate the use of NTI in patients undergoing oral cavity reconstruction with FTT. Our results suggest that in the appropriate institutional setting, most patients can be safely managed with NTI. This approach results in a decreased hospital LOS and ICU LOS and an earlier resumption of oral intake with less need for gastrostomy.