Understanding egress outflow of cerebrospinal fluid (CSF) is essential for exploring its role in brain health and its relationship to neurodegenerative diseases. Existing models often fail to capture the complex flow dynamics, particularly distinguishing rapid bulk flow from slower, perfusion-like components. This study aims to develop and validate a bi-component model that characterizes these distinct signal patterns using Time-Spatial Labeling Inversion Pulse (Time-SLIP) MRI data. A bi-component model is proposed by combining a Gaussian function (to represent fast bulk outflow) with a Г-variate function (representing slower, perfusion-like flow). Validation phantom studies were conducted using a sealed water phantom with computer-controlled bulk flow. In-vivo meninges images in eight healthy subjects were acquired on a clinical 3T MRI using Time-SLIP with 3D single-shot FSE. Signal Increase Ratio (SIR) from these experiments were fitted using the proposed bi-component model with fitting accuracy assessed using coefficient of determination (R2), Sum of Squared estimate of Errors (SSE), and Root Mean Square Error (RMSE). Phantom studies showed superior fitting of bulk fluid dynamics by the Gaussian component, confirmed by lower SSE, RMSE and higher R2. The in-vivo results demonstrated that the bi-component model provided a superior fit, capturing two distinct components of CSF outflow, which are hypothesized to correspond to fast bulk flow and a slower, perfusion-like component, which could be a mixture of CSF and interstitial fluid. The bi-component model enhances characterization of CSF egress outflow from intrinsic labeling by Time-SLIP, providing potential quantitative biomarkers for the assessment of neurodegenerative diseases.
Malis et al. (Thu,) studied this question.