RESUMEN
Background: Skin electrodes have been reported to be a useful alternative recording method for intraoperative neuromonitoring (IONM) and show typical electromyography (EMG) waveforms while overcoming the shortcomings of the EMG endotracheal tube. However, the skin electrodes showed relatively lower evoked amplitudes than other recording methods. In this study, we analyzed normative EMG data using skin electrodes and factors that affect the evoked amplitude of thyroid IONM. Methods: In total, 167 patients [242 nerves at risk (NAR)] who underwent thyroidectomy under IONM with adhesive skin electrodes were analyzed. A pair of skin electrodes was attached to the lateral border of the lamina of the thyroid cartilage. Evoked EMG data, including mean amplitude and latency, obtained after stimulation of the recurrent laryngeal nerve (RLN) and vagus nerve (VN), were collected and analyzed. Results: The mean amplitudes of RLN and VN recorded via skin electrodes were 255.48±96.53 and 236.15±69.72 µV, respectively. The mean latency of the right and left RLN was 3.22±0.03 and 3.49±0.08 mS, respectively. The mean latency of the right and left VN was 5.37±0.80 and 7.57±0.10 mS, respectively. The mean amplitude was significantly lower in the obesity, male, and total thyroidectomy (TT) groups. As body mass index (BMI) and age increased, the amplitude of EMG tended to decrease significantly. Conclusions: The evoked amplitude recorded with the skin electrodes was relatively low. A larger surgical extent, obesity, male sex, and age >55 years showed significantly lower evoked amplitudes.
RESUMEN
BACKGROUND: Many studies have been reported on tracheostomy to prevent upper airway obstruction after surgery. Among these, the scoring system proposed by Cameron et al. quantifies various factors that influence postoperative respiratory failure. This system provides a basis for surgeons to decide whether to perform an elective tracheostomy. In this study, the authors applied the Cameron scoring system retrospectively to patients undergoing severe oral cancer surgery to reevaluate the indications for elective tracheostomy and to investigate its clinical efficacy in airway management. In this study, a sample of 20 patients who underwent oral cancer surgery was selected and divided into two groups: 10 underwent tracheostomy and 10 did not. The Cameron scoring scores for each patient were extracted, to verify whether elective tracheostomy was performed in accordance with the threshold scores. Differences in scores and significant clinical impact factors between the two groups were analyzed and compared. RESULT: The 10 patients who underwent tracheostomy had an average Cameron score of 6.4, all scoring above the recommended threshold of 5 for tracheostomy. For the 10 patients who did not undergo tracheostomy, the average score was 2.5, with 8 out of these 10 patients scoring below 5. Significant clinical impact factors observed included the location and size of the tumor, the performance of mandibulectomy and neck dissection, and the type of reconstruction surgery. CONCLUSION: In planning surgery for oral cancer patients, it is essential to consider the use of elective tracheostomy based on preoperative assessment of the risk of postoperative airway obstruction using tools like the Cameron scoring system, and patients' condition. Research confirms that elective tracheostomy effectively enhances airway management in patients with severe oral cancer.
RESUMEN
OBJECTIVES: This study aimed to establish an artificial intelligence (AI) system to classify vertical level differences between vocal folds during vocalization and to evaluate the accuracy of the classification. METHODS: We designed models with different depths between the right and left vocal folds using an excised canine larynx. Video files for the data set were obtained using a high-speed camera system and a color complementary metal oxide semiconductor camera with global shutter. The data sets were divided into training, validation, and testing. We used 20,000 images for building the model and 8000 images for testing. To perform deep learning multiclass classification and to estimate the vertical level difference, we introduced DenseNet121-ConvLSTM. RESULTS: The model was trained several times using different numbers of epochs. We achieved the most optimal results at 100 epochs, and the batch size used during training was 16. The proposed DenseNet121-ConvLSTM model achieved classification accuracies of 99.5% and 88.0% for training and testing, respectively. After verification using an external data set, the overall accuracy, precision, recall, and f1-score were 90.8%, 91.6%, 90.9%, and 91.2%, respectively. CONCLUSIONS: The newly developed AI system may be an easy and accurate method for classifying superior and inferior vertical level differences between vocal folds. Thus, this AI system can be applied and may help in the assessment of vertical level differences in patients with unilateral vocal fold paralysis.