Magnetic susceptibility measurements from magnetic resonance imaging (MRI)

Detection of traces of iron in the brain is useful for early diagnosis, progression and treatment monitoring of various neurologic and psychiatric diseases. Iron and, in general, different tissues possessing different magnetic susceptibilities expose themselves via small spatial variations of the magnetic field during Magnetic Resonance Imaging (MRI). Data acquisition and image analysis methods that visualize spatial distributions of magnetic susceptibilities are known as Quantitative Susceptibility Mapping. However, most modern techniques model the brain as a collection of magnetic dipoles and because only the longitudinal component of magnetic field is assessable to MRI, the resultant problem is ill posed. To address this, we are developing a new tissue model and algorithm to map susceptibility from the experimentally measured magnetic field in the whole region. Our algorithm addresses this problem by first solving the forward magnetostatics problem of the sample inside the coils for a trial susceptibility map. Second, this solution is used to compare the theoretical longitudinal field component with the experimental one. Third, the susceptibility map is varied until the difference between the numerical and experimental field components is minimized.