DECAES - DEcomposition and Component Analysis of Exponential Signals

Area of Science:

• Biophysics
• Medical Imaging
• Computational Science

Background:

• Multi-exponential signals are prevalent in scientific disciplines, requiring decomposition for component analysis.
• Magnetic resonance imaging (MRI) generates multi-exponential T2* and T2 decays, posing computational challenges due to large datasets.
• Existing decomposition methods are time-consuming, hindering real-time applications like on-scanner clinical analysis.

Purpose of the Study:

• To develop a computationally efficient approach for decomposing multi-exponential signals.
• To accelerate the generation of quantitative imaging maps from MRI data.

Main Methods:

• A novel, fast algorithm for multi-exponential signal decomposition was developed.
• The method was applied to multi-echo spin echo MRI scans.

Main Results:

• The new approach significantly reduces computation time for signal decomposition.
• Whole-brain myelin water mapping was achieved in under 2 minutes.
• Prostate luminal water mapping was completed in under 1 minute.

Conclusions:

• The presented method offers a substantial speed improvement for multi-exponential signal decomposition in MRI.
• This fast decomposition enables rapid generation of myelin and luminal water maps, facilitating clinical applications.
• The approach addresses the computational bottleneck in analyzing complex MRI data.