Velocity Selective Arterial Spin Labeling Perfusion Imaging at 3T and 7T

Sai Abitha Srinivas - BME Master's Thesis Defense

WHERE: Gerstacker Building, 1117 show on map

WHEN: December 10, 2018 3:00 pm-4:00 pmADD TO CALENDAR

Velocity Selective Arterial Spin Labeling Perfusion Imaging at 3T and 7T: Sai Abitha Srinivas - BME Master's Thesis Defense

Purpose: Hemodynamic parameters, such as perfusion, are key indicators of organ function. Arterial Spin Labeling (ASL) allows us to perform perfusion measurements without injection of exogenous tracers. The Purpose of this thesis is to implement and improve a Fourier-Transform based Velocity Selective Inversion (FT-VSI) pulse train for Velocity-Selective Arterial Spin Labeling (VSASL) on a 7 T Scanner. Additionally, to perform Abdominal perfusion Imaging using VSASL on a 3T Scanner.



Methods: The FT-VSI was calibrated and simulated to check for velocity profiles. It's sensitivities to B0/B1 inhomogeneity and gradient imperfections such as eddy currents were evaluated through phantom studies. A flow phantom was used to test for inversion efficiency . At 3T, a velocity selective saturation(VSS) pulse was used to image Kidney and Spinal Cord blood perfusion. The tracer kinetic properties of VSS pulses for renal and spinal cord perfusion were characterized.



Results: Phantom results of the proposed FT-VSI pulse train demonstrated high correlation to B0/B1 field inhomogeneity. A high T2* decay was observed at 7T. Through Simulations, the FT-VSI was improved for higher velocity selectivity and shorter length of pulse to counteract this decay. Eddy current effects were highly controllable by introducing gaps between consecutive gradient pulses. At 3T, ASL images collected at various labeling delays after the VSS pulse. ASL values such as blood volume(BV), Blood Flow(BF), Bolus Arterial Transit time (ATT) and bolus width were estimated by fitting a two compartment models. The Kidney perfusion values (Medulla and cortex) were in agreement with literature values. Although Lower perfusion and blood volume in the spinal cord resulted in poor fits, Spinal cord flow noticeably did not experience a delay in the label arrival



Conclusion: A FT-VSI pulse train was demonstrated on a 7T to be a suitable labeling module for VSASL with robustness of velocity selective profile to gradient imperfections but not to B0/B1 field inhomogeneity. Application of velocity selective pulses for abdominal imaging were demonstrated at 3T.


Chair: Dr. Luis Hernandez-Garcia