Pulmonary Nodule Classification Using ML and DL Techniques for Lung Cancer Detection

Lung cancer continues to be a significant cause of mortality worldwide, primarily due to late-stage diagnosis and the challenges in distinguishing between benign and malignant pulmonary nodules. Recent advances in Machine Learning (ML) and Deep Learning (DL) have revolutionized medical diagnostics by enabling automated, accurate, and non-invasive classification of pulmonary nodules from CT images. This study proposes a hybrid approach that integrates Deep Convolutional Neural Networks (DCNNs) such as AlexNet, VGG-16, and VGG-19 with ensemble learning methods including Support Vector Machine (SVM) and AdaBoostM2. Ensemble fusion techniques like averaging and MAX-VOTE are applied to enhance classification accuracy and robustness. Experimental evaluations using the LIDC-IDRI and LUNA16 datasets demonstrate that the fusion-based ensemble learners significantly outperform single classifiers, achieving accuracy levels exceeding 96%. The results highlight the potential of hybrid ML-DL frameworks in improving early lung cancer detection and supporting computer-aided diagnosis systems.