Lung Nodule Segmentation using Deep Learning and Advanced UNet Model

Cancer is known as one of the world’s top reason mortality in human beings. Lung cancer, notably, has highest rate. Thus, timely detection nodule or tumor a critical and significant job saving lives. One hot topic current research field automatic lung nodules. Many methods have been implemented using computer vision-based technologies past, but achieving desired precision still remains difficult job. In this research, we adopt Convolutional Neural Network (CNN) based UNet image segmentation model improved its architecture by incorporating convolution mechanisms. Moreover, scheme uses binary cross entropy loss function during training process. The proposed mechanism tested on LIDC-IDRI dataset. experimental analysis shows augmented performance approach when compared with existing techniques. qualitative quantitative comparative that suggested substantially improves efficiency performance.