Incorporating prior shape knowledge via data-driven loss model to improve 3D liver segmentation in deep CNNs.

Journal: International journal of computer assisted radiology and surgery

PMID: 31686380

Abstract

PURPOSE: Convolutional neural networks (CNNs) have obtained enormous success in liver segmentation. However, there are several challenges, including low-contrast images, and large variations in the shape, and appearance of the liver. Incorporating prior knowledge in deep CNN models improves their performance and generalization.

Authors

Saeed Mohagheghi

Department of Biomedical Engineering, Engineering Faculty, Shahed University, Tehran, Iran.
Amir Hossein Foruzan

Biomedical Engineering Group, Engineering Faculty, Shahed University, Iran.

Keywords

Algorithms Humans Image Processing, Computer-Assisted Liver Models, Anatomic Neural Networks, Computer Tomography, X-Ray Computed

External Resources

View on PubMed Access via DOI PubMed (31686380)

Incorporating prior shape knowledge via data-driven loss model to improve 3D liver segmentation in deep CNNs.

Abstract

Authors

Keywords

External Resources

Popular Topics

Recent Journals