Machine Learning Based Prediction of Recurrence in Non-Small Cell Lung Cancer: A Multi Model Approach Integrating Clinical and Radiomic Features
Machine Learning Based Prediction of Recurrence in Non-Small Cell Lung Cancer: A Multi Model Approach Integrating Clinical and Radiomic Features
Mohammad Ayesha Summaiyya¹, T. Monika2, Dr. M. Chandran3, Dr. T. Kumanan4, Dr. M. Nisha5
Department of Computer Science and Engineering 1,2,3,4,5
Dr. M.G.R. Educational and Research Institute, Chennai 600095, India
Abstract:Background: Non-small cell lung cancer (NSCLC)makes up about 85% of lung cancers Cases worldwide recurrenceafter initial treatment is a key factor in poor long-term survival quickly and accurately identifying patients at high risk of recurrence is crucial for guiding treatments decisions and monitoring their progress after treatment. Methods: This study presents a machine learning framework that combines clinical variables and CT- derived radiomic features to predict NSCLC recurrence the analysis involves 422 patients from the cancer imaging archive (TCIA) lung 1 dataset. We extracted cli ical features including age gender, TNM staging, overall diseases stage and histological subtype, along with 59 radiomic descriptors, such as texture (GLCM), local binary patterns (LBP), wavelet transforms, and frequency domain (FFT) features. we evaluated four machine learning models: Logistic Regression, Random Forest, Gradient Boosting, and Deep Neural Network we used a weighted fusion strategy (clinical: 45% radiomic: 55%) for the multimodal framework.