
Although rare, salivary gland carcinomas (SGCs) are prevalent oral and maxillofacial malignancies. This study developed a machine learning-based model to predict survival outcomes for SGC patients. Researchers collected demographic and clinicopathological data from patients treated between 1963 and 2014. Using feature selection methods and multiple machine learning algorithms, they identified prognosis-related factors and created survival prediction models. The LightGBM algorithm demonstrated the highest performance, with an AUC of 0.83 and an accuracy of 0.91. This model offers the potential for personalized diagnostic and treatment strategies, improved doctor-patient communication, and individualized follow-up plans.
Like
Save
Share