Sparsity-regularized approaches to directly reconstructing hemodynamic response in brain functional diffuse optical...

Area of Science:

• Neuroimaging
• Biomedical Optics
• Medical Physics

Background:

• Brain functional diffuse optical tomography (fDOT) traditionally uses indirect methods to reconstruct absorption coefficients and then derive hemodynamic responses.
• Direct fDOT approaches combine these steps for improved efficiency but often yield low spatial resolution and blurred edges due to ill-posed inverse problems.

Purpose of the Study:

• To enhance the reconstruction quality in direct brain functional diffuse optical tomography.
• To address limitations of existing direct methods, specifically poor edge definition and low spatial resolution.

Main Methods:

• Introduction of a priori sparsity into the direct reconstruction framework for brain fDOT.
• Development of algorithms to leverage sparsity for obtaining sparse solutions.
• Conducting simulation experiments to validate the proposed approach.

Main Results:

• The proposed method, incorporating a priori sparsity, is expected to improve reconstruction quality.
• Sparse solutions lead to sharper edges and higher spatial resolution compared to conventional direct fDOT methods.
• Simulation experiments confirm the anticipated performance improvements.

Conclusions:

• Incorporating a priori sparsity is an effective strategy to enhance direct brain functional diffuse optical tomography.
• The developed approach offers a promising advancement for more accurate and detailed neuroimaging using fDOT.
• This method has the potential to improve the understanding of brain function through improved spatial resolution in hemodynamic imaging.