Abstract
Magnetic resonance imaging (MRI) offers the ability to non-invasively map the brain’s metabolic oxygen consumption (CMRO2), which is essential for understanding and monitoring neural function in both health and disease. One MRI method of mapping CMRO2 is based on the simultaneous acquisition of cerebral blood flow (CBF) and blood oxygen level dependent (BOLD) weighted images during respiratory modulation of both oxygen and carbon dioxide. Although this dual-calibrated methodology has shown promise in the research setting, current analysis methods are unstable in the presence of noise and/or are computationally demanding. In this paper, we present a machine learning implementation for the multi-parametric assessment of dual-calibrated fMRI data. An extremely randomized trees regressor and a multi-layer perceptron (MLP) are cascaded to provide quantitative estimates of the resting CBF, CMRO2, and oxygen extraction fraction (OEF). The proposed implementation takes advantage of the inherent noise immunity of tree-based ensemble methods and MLPs to provide robust and computationally efficient estimates of CBF, CMRO2, and OEF. Synthetic data with additive noise are used to train the regressors, and their performance is compared to conventional analysis methods both in simulation and with in-vivo data (n=30). The proposed method is demonstrated to significantly reduce computation time, error, and proportional bias in both CMRO2 and OEF estimates.
Footnotes
Figure 5 included