A REFINED VGG-19 MODEL FOR ACCURATE CLASSIFICATION OF LUNG CANCER

Lung cancer remains one of the leading causes of cancer deaths worldwide, so proper and effective screening methods are crucial. Machine learning algorithms have been shown to be very useful in medical diagnostics, as they can analyze very complex medical data with high accuracy. This work discusses an ML-based approach to lung cancer detection based on advanced feature extraction and classification techniques for the identification of cancerous patterns in medical images. By applying a comprehensive dataset of lung CT scan images, the proposed model outperforms traditional methods by remarkable accuracy, sensitivity, and specificity. With the assistance of predefined pre-processing steps and optimized algorithms of ML, the algorithm can work robustly against noise and variability in data. This work highlights the transformative potential of machine learning in early detection of lung cancer, leading to a better outcome and increased survival rate of patients. Various CNN architectures, including AlexNet, Inception-ResNet-V2, and VGG16, were evaluated to determine the most effective model for lung cancer detection. An enhanced VGG-19 approach emerged as the superior framework, achieving an impressive accuracy of approximately 99%, significantly outperforming other architectures.