Enhanced Efficiency in Lung Cancer Classification via Deep Learning Ensembles

Lung cancer, a prominent global cause of mortality, underscores the significance of early diagnosis for improved patient survival rates. The advent of Computer-Aided Detection (CAD) systems has emerged as a pivotal strategy in expediting lung cancer detection, potentially safeguarding lives. In the realm of artificial intelligence, deep learning has garnered increasing attention within the medical domain. This research leverages lung nodule images extracted from CT scans within the Lung 1 dataset, classifying them into adenocarcinoma and small-cell carcinoma categories. The feature selection employs the transfer learning algorithm VGG16, while the classification task is undertaken using the support vector machine (SVM) learning algorithm. Notably, this approach yields a commendable accuracy of 98%, underscoring its potential significance in the realm of early lung cancer detection and patient care.