Task-based assessment of a convolutional neural network for segmenting breast lesions for radiomic analysis.

Journal: Magnetic resonance in medicine
Published Date: Apr 8, 2019

Abstract

PURPOSE: Radiomics allows for powerful data-mining and feature extraction techniques to guide clinical decision making. Image segmentation is a necessary step in such pipelines and different techniques can significantly affect results. We demonstrate that a convolutional neural network (CNN) segmentation method performs comparably to expert manual segmentations in an established radiomics pipeline.

Authors

  • Karl D Spuhler
    Department of Biomedical Engineering, Stony Brook University, Stony Brook, New York.
  • Jie Ding
    State Key Laboratory of Respiratory Disease, Joint School of Life Sciences, Guangzhou Chest Hospital, Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences, Guangzhou Medical University, Guangzhou, China.
  • Chunling Liu
    Department of Applied Mathematics, Hong Kong Polytechnic University, Hong Kong, China.
  • Junqi Sun
    Radiology, Renaissance School of Medicine, Stony Brook University, Stony Brook, New York.
  • Mario Serrano-Sosa
    Biomedical Engineering, Stony Brook University, Stony Brook, New York.
  • Meghan Moriarty
    Radiology, Renaissance School of Medicine, Stony Brook University, Stony Brook, New York.
  • Chuan Huang
    Department of Biomedical Engineering, Stony Brook University, Stony Brook, New York chuan.huang@stonybrookmedicine.edu.

Keywords

External Resources

Popular Topics
Recent Journals