Related Experiment Video
Updated: Jun 8, 2025
Magnetic Resonance Imaging of Multiple Sclerosis at 7.0 Tesla
Published on: February 19, 2021
Perceptual super-resolution in multiple sclerosis MRI
Diana L Giraldo1,2,3, Hamza Khan4,5,6, Gustavo Pineda3
1Imec-Vision Lab, University of Antwerp, Antwerp, Belgium.
Introduction:
Magnetic resonance imaging (MRI) is crucial for diagnosing and monitoring of multiple sclerosis (MS) as it is used to assess lesions in the brain and spinal cord. However, in real-world clinical settings, MRI scans are often acquired with thick slices, limiting their utility for automated quantitative analyses. This work presents a single-image super-resolution (SR) reconstruction framework that leverages SR convolutional neural networks (CNN) to enhance the through-plane resolution of structural MRI in people with MS (PwMS).
Methods:
Our strategy involves the supervised fine-tuning of CNN architectures, guided by a content loss function that promotes perceptual quality, as well as reconstruction accuracy, to recover high-level image features.
Results:
Extensive evaluation with MRI data of PwMS shows that our SR strategy leads to more accurate MRI reconstructions than competing methods. Furthermore, it improves lesion segmentation on low-resolution MRI, approaching the performance achievable with high-resolution images.
Discussion:
Results demonstrate the potential of our SR framework to facilitate the use of low-resolution retrospective MRI from real-world clinical settings to investigate quantitative image-based biomarkers of MS.
Insights
This study introduces a new super-resolution (SR) method using convolutional neural networks (CNNs) to improve low-resolution MRI scans for multiple sclerosis (MS) patients. The technique enhances image quality and lesion detection in MS, aiding quantitative biomarker analysis.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Neurology
Background:
- Magnetic resonance imaging (MRI) is vital for multiple sclerosis (MS) diagnosis and monitoring.
- Clinical MRI scans often use thick slices, hindering automated quantitative analysis.
- Improving the resolution of retrospective MRI data is crucial for MS research.
Purpose of the Study:
- To develop a super-resolution (SR) reconstruction framework for enhancing through-plane resolution of structural MRI in people with MS (PwMS).
- To leverage SR convolutional neural networks (CNNs) for improving low-resolution MRI quality.
- To enable quantitative analysis of retrospective clinical MRI data.
Main Methods:
- Employed supervised fine-tuning of CNN architectures.
- Utilized a content loss function to enhance perceptual quality and reconstruction accuracy.
- Focused on recovering high-level image features for improved resolution.
Main Results:
- The proposed SR strategy yielded more accurate MRI reconstructions compared to existing methods.
- Significantly improved lesion segmentation on low-resolution MRI scans.
- Achieved performance comparable to high-resolution images for lesion detection.
Conclusions:
- The SR framework enhances the utility of retrospective, low-resolution clinical MRI for PwMS.
- This approach facilitates the investigation of quantitative image-based biomarkers for MS.
- The method holds potential for broader application in MS research and clinical practice.
