Improving MRI differentiation of gray and white matter in epileptogenic lesions based on nonlinear feedback.

A new method for enhancing MRI contrast between gray matter (GM) and white matter (WM) in epilepsy surgery patients with symptomatic lesions is presented. This method uses the radiation damping feedback interaction in high-field MRI to amplify contrast due to small differences in resonance frequency in GM and WM corresponding to variations in tissue susceptibility. High-resolution radiation damping-enhanced (RD) images of in vitro brain tissue from five patients were acquired at 14 T and compared with corresponding conventional T(1)-, T(2) (*)-, and proton density (PD)-weighted images. The RD images yielded a six times better contrast-to-noise ratio (CNR = 44.8) on average than the best optimized T(1)-weighted (CNR = 7.92), T(2) (*)-weighted (CNR = 4.20), and PD-weighted images (CNR = 2.52). Regional analysis of the signal as a function of evolution time and initial pulse flip angle, and comparison with numerical simulations confirmed that radiation damping was responsible for the observed signal growth. The time evolution of the signal in different tissue regions was also used to identify subtle changes in tissue composition that were not revealed in conventional MR images. RD contrast is compared with conventional MR methods for separating different tissue types, and its value and limitations are discussed.