Densely Connected Recurrent Residual (Dense R2UNet) Convolutional Neural Network for Segmentation of Lung CT Images

TL;DR

This paper introduces Dense R2U CNN, a novel deep learning architecture combining recurrent, residual, and dense layers within a U-Net framework, achieving improved lung CT image segmentation.

Contribution

The paper presents a new Dense R2U CNN model that integrates recurrent, residual, and dense connections for enhanced segmentation performance in medical imaging.

Findings

Outperforms equivalent models on lung lesion segmentation

Improves feature propagation and training depth

Achieves better accuracy on benchmark dataset

Abstract

Deep Learning networks have established themselves as providing state of art performance for semantic segmentation. These techniques are widely applied specifically to medical detection, segmentation and classification. The advent of the U-Net based architecture has become particularly popular for this application. In this paper we present the Dense Recurrent Residual Convolutional Neural Network(Dense R2U CNN) which is a synthesis of Recurrent CNN, Residual Network and Dense Convolutional Network based on the U-Net model architecture. The residual unit helps training deeper network, while the dense recurrent layers enhances feature propagation needed for segmentation. The proposed model tested on the benchmark Lung Lesion dataset showed better performance on segmentation tasks than its equivalent models.