Direct estimation of evoked hemoglobin changes by multimodality fusion imaging

Area of Science:

• Neuroimaging
• Biophysics
• Biomedical Engineering

Background:

• Diffuse optical tomography (DOT) and blood oxygen level dependent (BOLD)-based functional magnetic resonance imaging (fMRI) are noninvasive techniques for brain activity monitoring.
• Both DOT and fMRI measure cerebral hemodynamic changes from hemoglobin variations but rely on different physical principles, presenting unique strengths and limitations.
• Integrating these modalities could leverage their complementary resolutions for enhanced brain activity analysis.

Purpose of the Study:

• To develop a unified linear model for combining concurrently measured DOT and fMRI signals.
• To enable cross-calibrated estimation of absolute deoxyhemoglobin changes using multimodal neuroimaging.
• To assess the calibration of the 3 tesla BOLD signal against deoxyhemoglobin concentrations.

Main Methods:

• Development of a unified linear model to integrate DOT and fMRI data.
• Numerical simulations to validate the model's capability for cross-calibrated deoxyhemoglobin estimation.
• Application of the multimodal analysis to experimental data acquired during a motor task.

Main Results:

• Concurrent DOT and BOLD measurements allow for cross-calibrated estimates of absolute micromolar deoxyhemoglobin changes.
• The multimodal approach provides more robust hemoglobin change estimations compared to DOT alone.
• The study estimated the calibration of the 3 tesla BOLD signal to be -0.55%+/-0.40% signal change per micromolar change of deoxyhemoglobin.

Conclusions:

• A unified linear model effectively combines DOT and fMRI for improved brain activity imaging.
• Multimodal neuroimaging enhances the accuracy and robustness of hemodynamic change quantification.
• This approach provides essential cross-calibration for BOLD fMRI, advancing quantitative brain function studies.