Lung segmentation with NASNet-Large-Decoder Net

TL;DR

This paper introduces a novel lung segmentation model using NASNet-Large as an encoder with a decoder architecture, achieving high accuracy and outperforming existing methods in lung image analysis.

Contribution

The work presents a new NASNet-Large-based encoder-decoder architecture with a post-processing layer for improved lung segmentation accuracy.

Findings

Achieved 0.92 dice score in lung segmentation.

Outperformed state-of-the-art segmentation methods.

Effective post-processing improves segmentation quality.

Abstract

Lung cancer has emerged as a severe disease that threatens human life and health. The precise segmentation of lung regions is a crucial prerequisite for localizing tumors, which can provide accurate information for lung image analysis. In this work, we first propose a lung image segmentation model using the NASNet-Large as an encoder and then followed by a decoder architecture, which is one of the most commonly used architectures in deep learning for image segmentation. The proposed NASNet-Large-decoder architecture can extract high-level information and expand the feature map to recover the segmentation map. To further improve the segmentation results, we propose a post-processing layer to remove the irrelevant portion of the segmentation map. Experimental results show that an accurate segmentation model with 0.92 dice scores outperforms state-of-the-art performance.