The research focuses on classifying lung cancer using the VGG-19 architecture. The datasets were sourced from Iraq-Oncology Teaching Hospital with 70% of the data allocated for training and 30% for testing. Performance metrics were computed to evaluate the effectiveness of the classification method. Python is utilized for designing the algorithm and executed using Goggle Colab. The lung tumor classification using VGG-19 offers an accuracy of 95%, sensitivity of 88.79%, specificity of 98.25 %, and F1-Score of 93.28%. However, the low sensitivity value indicates that the VGG-19 architecture is not accurately predicting benign and malignant cases.

VGG-19 CNN Lung Tumor Accuracy Loss