Recurrent residual U-Net with EfficientNet encoder for medical image segmentation

Nahian Siddique; Sidike Paheding; Md. Zahangir Alom; Vijay Devabhaktuni

doi:10.1117/12.2591343

Translator Disclaimer

12 April 2021 Recurrent residual U-Net with EfficientNet encoder for medical image segmentation

Nahian Siddique, Sidike Paheding, Md. Zahangir Alom, Vijay Devabhaktuni

Author Affiliations +

Proceedings Volume 11735, Pattern Recognition and Tracking XXXII; 117350L (2021) https://doi.org/10.1117/12.2591343
Event: SPIE Defense + Commercial Sensing, 2021, Online Only

Abstract

In this paper, we propose a U-net architecture that integrates a residual skip connections and recurrent feedback with EfficientNet as a pretrained encoder. Residual connections help feature propagation in deep neural networks and significantly improve performance against networks with a similar number of parameters while recurrent connections ameliorate gradient learning. We also propose a second model that utilizes densely connected layers aiding deeper neural networks. EfficientNet is a family of powerful pretrained encoders that streamline neural network design. The proposed networks are evaluated against state-of-the-art deep learning based segmentation techniques to demonstrate their superior performance.

Conference Presentation

Citation Download Citation

Nahian Siddique, Sidike Paheding, Md. Zahangir Alom, and Vijay Devabhaktuni "Recurrent residual U-Net with EfficientNet encoder for medical image segmentation", Proc. SPIE 11735, Pattern Recognition and Tracking XXXII, 117350L (12 April 2021); https://doi.org/10.1117/12.2591343

ACCESS THE FULL ARTICLE