Related Experiment Video
Updated: Feb 18, 2026
10:46
Diffusion Imaging in the Rat Cervical Spinal Cord
Published on: April 7, 2015
12.3K
Identification of ghost artifact using texture analysis in pediatric spinal cord diffusion tensor images
Mahdi Alizadeh1, Chris J Conklin2, Devon M Middleton2
1Jefferson Integrated Magnetic Resonance Imaging Center, Department of Radiology, Thomas Jefferson University, Philadelphia, PA, United States; Department of Neurosurgery, Thomas Jefferson University, Philadelphia, PA, United States.
Magnetic Resonance Imaging
|November 21, 2017
Summary
This study developed an automated pipeline to remove ghost artifacts in pediatric spinal cord diffusion tensor imaging (DTI). The method achieved 84% accuracy, improving image quality for better diagnostics.
Area of Science:
- Medical Imaging
- Biomedical Engineering
- Neuroscience
Background:
- Ghost artifacts significantly degrade spinal cord diffusion tensor imaging (DTI) quality.
- Accurate artifact removal is essential for reliable spinal cord DTI analysis, especially in pediatric populations.
Purpose of the Study:
- To design, implement, and validate a multi-stage post-processing pipeline for automatic ghost artifact removal in pediatric spinal cord DTI.
- To improve the diagnostic utility of DTI by enhancing image clarity and reducing motion-related artifacts.
Main Methods:
- A cohort of 12 pediatric subjects (healthy and with spinal cord injury) was studied.
- Region of interests (ROIs) of ghost/true cords were segmented using mathematical morphological processing.
- Texture features were extracted and selected based on mutual information, then classified using an adaptive neuro-fuzzy interface system.
Main Results:
- The pipeline successfully separated ghost artifacts from true cord structures.
- The classifier achieved 91% sensitivity, 79% specificity, and 84% accuracy in distinguishing true cords from ghost artifacts.
Conclusions:
- The proposed method shows promise for automatic ghost artifact detection in spinal cord DTI.
- This automated approach is a crucial step towards developing robust post-processing pipelines for DTI analysis.
