Construction of prediction model of early glottic cancer based on machine learning

The early diagnosis of glottic laryngeal cancer is the key to successful treatment, and machine learning (ML) combined with narrow-band imaging (NBI) laryngoscopy provides a new idea for the early diagnosis of glottic laryngeal cancer.

To explore the clinical applicability of the diagnosis of early glottic cancer based on ML combined with NBI.

A retrospective study was conducted on 200 patients diagnosed with laryngeal mass, and the general clinical characteristics and pathological results of the patients were collected. Chi-square test and multivariate logistic regression analysis were used to explore clinical and laryngoscopic features that could potentially predict early glottic cancer. Afterward, three classical ML methods, namely random forest (RF), support vector machine (SVM), and decision tree (DT), were combined with NBI endoscopic images to identify risk factors related to glottic cancer and to construct and compare the predictive models.

The RF‑based model was found to predict more accurately than other methods and have a significant predominance over others. The accuracy, precision, recall and F1 index, and AUC value of the RF model were 0.96, 0.90, 1.00, 0.95, and 0.97.

We developed a prediction model for early glottic cancer using RF, which outperformed other models.