An Efficient DA-Net Architecture for Lung Nodule Segmentation

A typical growth of cells inside tissue is normally known as a nodular entity. Lung nodule segmentation from computed tomography (CT) images becomes crucial for early lung cancer diagnosis. An issue that pertains to the nodules homogenous modular variants. The resemblance among well neighboring regions very challenging deal with. Here, we propose an end-to-end U-Net-based framework named DA-Net efficient segmentation. This method extracts rich features by integrating compactly and densely linked convolutional blocks merged with Atrous convolutions broaden view filters without dropping loss coverage data. We first extract lung’s ROI whole CT scan slices using standard image processing operations k-means clustering. reduces search space model only lungs where are present instead slice. evaluation suggested was performed through utilizing LIDC-IDRI dataset. According results, found showed good performance, achieving 81% Dice score value 71.6% IOU score.