A nested parallel multiscale convolution for cerebrovascular segmentation

Likun Xia, Yixuan Xie, Qiwang Wang, Hao Zhang, Cheng He, Xiaonan Yang, Jinghui Lin, Ran Song, Jiang Liu, Yitian Zhao

Research output: Journal PublicationArticlepeer-review

Abstract

Purpose: Cerebrovascular segmentation in magnetic resonance imaging (MRI) plays an important role in the diagnosis and treatment of cerebrovascular diseases. Many segmentation frameworks based on convolutional neural networks (CNNs) or U-Net-like structures have been proposed for cerebrovascular segmentation. Unfortunately, the segmentation results are still unsatisfactory, particularly in the small/thin cerebrovascular due to the following reasons: (1) the lack of attention to multiscale features in encoder caused by the convolutions with single kernel size; (2) insufficient extraction of shallow and deep-seated features caused by the depth limitation of transmission path between encoder and decoder; (3) insufficient utilization of the extracted features in decoder caused by less attention to multiscale features. Methods: Inspired by U-Net++, we propose a novel 3D U-Net-like framework termed Usception for small cerebrovascular. It includes three blocks: Reduction block, Gap block, and Deep block, aiming to: (1) improve feature extraction ability by grouping different convolution sizes; (2) increase the number of multiscale features in different layers by grouping paths of different depths between encoder and decoder; (3) maximize the ability of decoder in recovering multiscale features from Reduction and Gap block by using convolutions with different kernel sizes. Results: The proposed framework is evaluated on three public and in-house clinical magnetic resonance angiography (MRA) data sets. The experimental results show that our framework reaches an average dice score of 69.29%, 87.40%, 77.77% on three data sets, which outperform existing state-of-the-art methods. We also validate the effectiveness of each block through ablation experiments. Conclusions: By means of the combination of Inception-ResNet and dimension-expanded U-Net++, the proposed framework has demonstrated its capability to maximize multiscale feature extraction, thus achieving competitive segmentation results for small cerebrovascular.

Original languageEnglish
Pages (from-to)7971-7983
Number of pages13
JournalMedical Physics
Volume48
Issue number12
DOIs
Publication statusPublished - Dec 2021
Externally publishedYes

Keywords

  • U-Net++
  • cerebrovascular segmentation
  • multiscale feature extraction

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging

Fingerprint

Dive into the research topics of 'A nested parallel multiscale convolution for cerebrovascular segmentation'. Together they form a unique fingerprint.

Cite this