Related Experiment Video
Updated: Jun 27, 2025
14:27
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
15.6K
Comparison between R2'-based and R2*-based χ-separation methods: A clinical evaluation in individuals with multiple
Sooyeon Ji1, Jinhee Jang2, Minjun Kim1
1Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea.
NMR in Biomedicine
|May 2, 2024
Summary
Two new R2*-based susceptibility source separation methods show promise for brain imaging. The deep learning approach, χ-sepnet-R2*, closely matches traditional methods, offering a viable alternative with reduced scan times for multiple sclerosis research.
Area of Science:
- Neuroimaging
- Magnetic Resonance Imaging (MRI)
- Quantitative Susceptibility Mapping
Background:
- Susceptibility source separation (χ-separation) estimates brain's diamagnetic (χdia) and paramagnetic (χpara) susceptibility using R2' maps.
- New R2*-based methods enable χ-separation using only multi-echo gradient echo (ME-GRE) data, reducing scan time and improving clinical utility.
- The impact of omitting R2 information in R2*-based methods requires evaluation.
Purpose of the Study:
- To evaluate the viability of two R2*-based χ-separation methods (model-based R2*-χ-separation and deep learning-based χ-sepnet-R2*) as alternatives to their R2'-based counterparts.
- To compare their performance in individuals with multiple sclerosis (MS) using qualitative and quantitative analyses.
Main Methods:
- Two R2*-based χ-separation methods (R2*-χ-separation and χ-sepnet-R2*) were compared against their R2'-based counterparts (χ-separation and χ-sepnet-R2').
- Performance was assessed in MS patients via qualitative visual assessment by neuroradiologists and quantitative analyses (ROI, linear regression).
- Comparisons focused on diamagnetic (χdia) and paramagnetic (χpara) susceptibility values and susceptibility offsets.
Main Results:
- Qualitatively, χ-sepnet-R2* closely aligned with χ-sepnet-R2', while R2*-χ-separation showed less distinct lesion contrasts.
- Quantitative analysis revealed robust correlations (r ≥ 0.88) between R2*-based and R2'-based methods in both whole-brain and MS lesion regions.
- χ-sepnet-R2* demonstrated better linearity and negligible offsets compared to R2*-χ-separation, which exhibited larger offsets potentially indicating false positives for myelin or iron.
Conclusions:
- Both R2*-based χ-separation methods are viable alternatives to R2'-based approaches, offering reduced scan times.
- χ-sepnet-R2* provides superior alignment with its R2'-based counterpart and minimal susceptibility offsets, making it a promising tool.
- R2*-χ-separation's larger offsets warrant caution, as they may lead to misinterpretation of myelin or iron content in MS lesions.
Related Concept Videos
01:23Test for Homogeneity
2.0K
The goodness–of–fit test can be used to decide whether a population fits a given distribution, but it will not suffice to decide whether two populations follow the same unknown distribution. A different test, called the test for homogeneity, can be used to conclude whether two populations have the same distribution. To calculate the test statistic for a test for homogeneity, follow the same procedure as with the test of independence. The hypotheses for the test for homogeneity can...
2.0K
01:17Sign Test for Matched Pairs
131
The sign test for matched pairs offers a robust method for comparing two paired samples, often for the effects of an intervention in one of them. This method is very useful in situations where the underlying distribution of the data is unknown. The test compares two related samples—often pre- and post-treatment measurements on the same subjects—to determine if there are significant differences in their median values.
To conduct the sign test, we first calculate the differences in...
To conduct the sign test, we first calculate the differences in...
131