Related Experiment Video
Updated: Jan 24, 2026

Implementation of a Hyperbolic Vortex Plasma Reactor for the Removal of Micropollutants in Water
Published on: July 25, 2025
Water removal in MR spectroscopic imaging with L2 regularization
Liangjie Lin1,2, Michal Považan1, Adam Berrington1
1Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, Maryland.
A new L2-regularization method effectively removes unwanted water signals from human brain proton MR spectroscopic imaging (MRSI) data. This technique improves metabolite mapping accuracy by minimizing water signal contamination.
Area of Science:
- Magnetic Resonance Imaging
- Spectroscopy
- Medical Physics
Background:
- Proton MR spectroscopic imaging (MRSI) is crucial for non-invasively assessing brain metabolites.
- Residual or unsuppressed water signals can contaminate MRSI data, complicating metabolite quantification.
- Existing methods for water signal removal may have limitations in efficiency or accuracy.
Purpose of the Study:
- To introduce and evaluate a novel L2-regularization based postprocessing method for removing residual water signals in human brain MRSI.
- To compare the performance of the proposed method against existing water suppression techniques.
- To assess the impact of water signal removal on the quality of reconstructed metabolite maps.
Main Methods:
- The proposed method utilizes L2 regularization with a synthesized water-basis matrix designed to be orthogonal to metabolite signals.
- Simulated MRSI spectra with varying water amplitudes and in vivo human brain MRSI datasets were employed for validation.
- Performance was evaluated by comparing results with two established postprocessing methods for water signal removal.
Main Results:
- The L2 regularization method accurately estimated metabolite signals in simulated data, closely matching true values.
- Residual water signals were efficiently suppressed in both short- and long-echo time in vivo brain MRSI datasets.
- The method enabled the reconstruction of high-quality metabolite maps with significantly reduced water contamination.
- A notable difference in creatine signals was observed between datasets with and without water saturation, linked to magnetization transfer effects.
Conclusions:
- The L2-regularization method, using a synthesized water matrix derived from prior knowledge, proves effective for water signal removal in human brain MRSI.
- This approach enhances the reliability of metabolite quantification and mapping in clinical and research applications.
- The findings highlight the potential of L2 regularization as a robust tool for improving MRSI data quality.
Related Concept Videos
States of Water
Water freezes when the intermolecular forces are greater than the kinetic energy. Unlike most other substances, water is less dense in its solid state than in its liquid state. This is because each water molecule can form...
Water and Mineral Acquisition
Quality of Water
The Water Cycle
Effect of Sea Water on Concrete
Concrete in areas between tide marks,...
Regulation of Water Intake

