Related Experiment Video
Updated: Jul 16, 2026
07:59
Fat-Water Phantoms for Magnetic Resonance Imaging Validation: A Flexible and Scalable Protocol
Published on: September 7, 2018
Automated analysis of multi site MRI phantom data for the NIHPD project
Luke Fu1, Vladimir Fonov, Bruce Pike
1McConnell Brain Imaging center, Montreal Neurological Institute, Quebec, Canada. lukelpfu@yahoo.com
Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|March 16, 2007
Summary
This study analyzed variations in medical imaging data across multiple centers and over time. Results show significant differences in signal to noise ratio (SNR) and percent integral uniformity (PIU) between sites, impacting image quality metrics.
Area of Science:
- Medical Imaging Analysis
- Quantitative Imaging
- Radiology Quality Assurance
Background:
- Large multi-center studies require robust methods to quantify data variations.
- Inter-site and longitudinal variability can affect the reliability of imaging data.
- Standardized phantom scans are crucial for assessing imaging system performance.
Purpose of the Study:
- To investigate inter-site variability in key imaging metrics using American College of Radiology (ACR) phantom scans.
- To analyze longitudinal variations in these metrics over time.
- To identify factors contributing to observed data drifts.
Main Methods:
- Utilized ACR phantom scans from the NIHPD project for automated measurements.
- Quantified signal to noise ratio (SNR), percent integral uniformity (PIU), diameter, and height.
- Analyzed data for statistical differences across multiple sites and over time.
Main Results:
- Statistically significant differences in mean SNR and PIU were observed across sites.
- Maximum mean diameter difference between sites was 2 mm (1.1%); maximum mean height difference was 2.5 mm (1.7%).
- Observed an average drift of 0.4 mm/year for diameter and 0.5 mm/year for height, with trends dependent on site, modality, and manufacturer.
Conclusions:
- Inter-site variability significantly impacts quantitative imaging metrics in multi-center studies.
- Longitudinal drifts in phantom measurements are measurable and influenced by site-specific factors.
- Automated analysis of phantom scans is essential for monitoring and ensuring consistent imaging performance.
