EnRaFS: An Ensemble Ranking-Based Feature Selection Approach for Grading Gallbladder Cancer Using Radiomic Analysis

Grading of gallbladder cancer (GBC) is pivotal for the diagnosis and treatment planning of patients suffering from this disease. Radiomics has emerged as a non-invasive, imperative, and efficient way for disease diagnosis and prediction with the use of machine learning approaches on medical data. Given the large dimensionality of the data, it is important to choose the most significant features to aid in improved classification of patients with respect to the subtypes/grades of GBC. This paper proposes a novel ensemble ranking-based approach called EnRaFS, for feature selection to grade GBC patients’ using CT scan images. It combines the results of multiple feature selection methods to improve the accuracy of the ranking. The ranked features are then used to train the machine learning model to predict the grade of the cancer. The proposed approach has been evaluated on a dataset of 105 patients diagnosed with GBC and compared with other state-of-the-art feature selection methods based on accuracy measure. Our study concludes that the proposed approach can be used as an effective tool for grading GBC, which can help clinicians to make more informed decisions about the treatment of the disease.