Research on predictive model for tracheal tube sizes in adult double-lumen endotracheal intubation based on radiomics and artificial intelligence.

Objective: This study aims to develop a predictive model for tracheal tube sizes in adult double-lumen endotracheal intubation using radiomics and artificial intelligence (AI) technologies to enhance the safety and efficiency of intubation procedures. Methods: A retrospective study design was adopted. Computed tomography (CT) imaging data of the neck and chest from 500 adult patients were collected, and radiomic features were extracted. After a rigorous screening, 390 patients were included in the analysis. Radiomics techniques were applied to analyze CT images and extract features related to tracheal tube size selection. Predictive models were constructed using AI algorithms, including random forests, decision tree, support vector machines, and Baidu Wenxin ERNIE. Major results: Among the models constructed, the Baidu Wenxin ERNIE model exhibited the best predictive performance, achieving an accuracy of 0.77 on the test set. Primary evaluation metrics, including accuracy, precision, recall, and F1-score, were compared to determine the optimal predictive model. Conclusions: This study successfully developed a predictive model for tracheal tube sizes in adult double-lumen endotracheal intubation based on radiomics and AI, demonstrating high predictive accuracy. This model has the potential to provide clinicians with a convenient, rapid, and efficient method of airway assessment, thereby enhancing the safety and efficiency of double-lumen endotracheal intubation.