Automatic liver and tumour segmentation from CT images using Deep learning algorithm

https://doi.org/10.1016/j.rico.2021.100087Get rights and content
Under a Creative Commons license
Open access

Abstract

The diagnosis and treatment of liver diseases from computed tomography (CT) images is an indispensable task for segmentation of Liver & its tumours. Due to the uneven presence, fuzzy borders, diverse densities, shapes and sizes of lesions segmentation of liver & its tumour is a difficult task. At this point we mainly focused on deep learning algorithms for segmenting liver and its tumour from abdominal CT scan images thereafter minimising the time & energy used for a liver diseases diagnosis. The algorithm is used here is based on the modified ResUNet architecture. Here we present, an automatic method based on semantic segmentation convolutional neural networks (CNNs) to segment Liver from CT scans and lesions from segmented liver part. The proposed system attains a Dice Similarity Coefficients (DSCs) of 96.35% and 89.28% and accuracy of 99.71% and 99.72% for liver and tumour segmentations, respectively. Comparison with the linked methods confirms the promise of the proposed system for liver and tumour segmentations.

Keywords

Liver & tumour segmentation
CNN
Deep learning
CT image
ResUNet
DSC