University of Oulu

Nguyen, D. H. M., Nguyen, D. M., Mai, T. T. N., Nguyen, T., Tran, K. T., Nguyen, A. T., Pham, B. T., & Nguyen, B. T. (2022). ASMCNN: An efficient brain extraction using active shape model and convolutional neural networks. Information Sciences, 591, 25–48.

ASMCNN : an efficient brain extraction using active shape model and convolutional neural networks

Saved in:
Author: Nguyen, Duy H. M.1; Nguyen, Duy M.2; Mai, Truong T. N.3;
Organizations: 1Max Planck Institute for Informatics, Germany
2Dublin City University, Ireland
3Department of Multimedia Engineering, Dongguk University, South Korea
4Department of Mathematics, University of Louisiana at Lafayette, USA
5Center for Machine Vision and Signal Analysis, University of Oulu, Finland
6Department of Mathematics, University of Architecture of Ho Chi Minh City, Viet Nam
7Department of Computer Science, Saigon University, Ho Chi Minh City, Viet Nam
8Department of Computer Science, VNU-HCMUS, Ho Chi Minh City, Viet Nam
9Vietnam National University, Ho Chi Minh City, Viet Nam
Format: article
Version: accepted version
Access: embargoed
Persistent link:
Language: English
Published: Elsevier, 2022
Publish Date: 2024-01-19


Brain extraction (skull stripping) is a challenging problem in neuroimaging. It is due to the variability in conditions from data acquisition or abnormalities in images, making brain morphology and intensity characteristics changeable and complicated. In this paper, we propose an algorithm for skull stripping in Magnetic Resonance Imaging (MRI) scans, namely ASMCNN, by combining the Active Shape Model (ASM) and Convolutional Neural Network (CNN) for taking full of their advantages to achieve remarkable results. Instead of working with 3D structures, we process 2D image sequences in the sagittal plane. First, we divide images into different groups such that, in each group, shapes and structures of brain boundaries have similar appearances. Second, a modified version of ASM is used to detect brain boundaries by utilizing prior knowledge of each group. Finally, CNN and post-processing methods, including Conditional Random Field (CRF), Gaussian processes, and several special rules are applied to refine the segmentation contours. Experimental results show that our proposed method outperforms current state-of-the-art algorithms by a significant margin in all experiments.

see all

Series: Information sciences
ISSN: 0020-0255
ISSN-E: 1872-6291
ISSN-L: 0020-0255
Volume: 591
Pages: 25 - 48
DOI: 10.1016/j.ins.2022.01.011
Type of Publication: A1 Journal article – refereed
Field of Science: 113 Computer and information sciences
213 Electronic, automation and communications engineering, electronics
Copyright information: © 2022. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http:/