Manara - Qatar Research Repository
10.1038_s41598-022-16828-6.pdf (2.34 MB)

A lightweight neural network with multiscale feature enhancement for liver CT segmentation

Download (2.34 MB)
journal contribution
submitted on 2024-04-25, 06:42 and posted on 2024-04-25, 06:42 authored by Mohammed Yusuf Ansari, Yin Yang, Shidin Balakrishnan, Julien Abinahed, Abdulla Al-Ansari, Mohamed Warfa, Omran Almokdad, Ali Barah, Ahmed Omer, Ajay Vikram Singh, Pramod Kumar Meher, Jolly Bhadra, Osama Halabi, Mohammad Farid Azampour, Nassir Navab, Thomas Wendler, Sarada Prasad Dakua

Segmentation of abdominal Computed Tomography (CT) scan is essential for analyzing, diagnosing, and treating visceral organ diseases (e.g., hepatocellular carcinoma). This paper proposes a novel neural network (Res-PAC-UNet) that employs a fixed-width residual UNet backbone and Pyramid Atrous Convolutions, providing a low disk utilization method for precise liver CT segmentation. The proposed network is trained on medical segmentation decathlon dataset using a modified surface loss function. Additionally, we evaluate its quantitative and qualitative performance; the Res16-PAC-UNet achieves a Dice coefficient of 0.950 ± 0.019 with less than half a million parameters. Alternatively, the Res32-PAC-UNet obtains a Dice coefficient of 0.958 ± 0.015 with an acceptable parameter count of approximately 1.2 million.

Publisher Correction: A lightweight neural network with multiscale feature enhancement for liver CT segmentation:, published online 21 September 2022.

Other Information

Published in: Scientific Reports
See article on publisher's website:



  • English


Springer Nature

Publication Year

  • 2022

License statement

This Item is licensed under the Creative Commons Attribution 4.0 International License.

Institution affiliated with

  • Hamad Medical Corporation
  • Hamad General Hospital - HMC
  • Hamad Bin Khalifa University
  • College of Science and Engineering - HBKU
  • Qatar University
  • Center for Advanced Materials - QU
  • College of Engineering - QU