Radiomic Feature Selection Using Gradient Loss of Deep Neural Network for Lung Cancer Stage Detection
Researchers have developed a new framework called Gradient-Loss Recursive Feature Elimination (GL-RFE) to improve the selection of radiomic features for lung cancer stage detection. This method utilizes the gradient sensitivity of a deep neural network to identify the most influential features from high-dimensional medical imaging data. The GL-RFE framework successfully identified 15 key features from chest CT scans, which were then used to train a deep neural network classifier, achieving a high accuracy of 90.22% in distinguishing between early and advanced lung cancer stages. AI
IMPACT Enhances the accuracy of AI-driven medical diagnostics by improving feature selection for high-dimensional imaging data.