Lung cancer is a leading cause of death globally and early detection is crucial for successful treatment. Deep learning with image processing techniques has become a popular and effective tool for medical image analysis, particularly in the detection and segmentation of lung cancer. This survey paper provides a comprehensive overview of the recent advancements in the use of DL (Deep Learning) for lung cancer detection and segmentation. The aim of this study was to compare the accuracy and sensitivity of three different image segmentation methods for lung cancer detection and segmentation. The methods evaluated in the study were DB U-Net+LLIE, U-net and DenseNet & dilation block with U-Net. The study found that DenseNets&dilation block with U-Net achieved the highest accuracy and sensitivity rates of 95.05% and 90.52%, respectively.Accurate segmentation is crucial for reliable diagnosis and treatment planning in lung cancer using histopathological images, and advanced deep learning methods such as DenseNets & dilation block with U-Net should be used to achieve higher accuracy levels.

Deep Learning, Image processing, Lung cancer, Histopathilogical images, Prediction