Faculty and Staff

Devashish Shrivastava, PhD Assistant Professor 612-626-2001 dev@cmrr.umn.edu

Dr. Shrivastava received his Ph.D. from the department of Mechanical Engineering at the University of Utah in 2005. His doctoral research, conducted under the mentorship of Professor Robert Roemer, focused on developing a blood vasculature based bioheat thermal model to predict in vivo temperature distribution in perfused tissues heated with external and/or internal heat sources. After receiving his Ph.D., Dr. Shrivastava joined the CMRR as a post-doctoral associate of Professor J. Thomas Vaughan. During the two years of his post-doctoral training, Dr. Shrivastava learned about the radiofrequency (RF) heating and other safety issues related to human imaging at ultra-high fields (i.e., MR systems with field strength greater than or equal to 3 tesla (T)). Currently, Dr. Shrivastava works as an Assistant Professor under the faculty mentorship of Professor J. Thomas Vaughan in the CMRR. Dr. Shrivastava studies RF heating, its thermo-physiological consequences and associated mechanisms to ensure RF safety for humans with and without conductive medical implants during imaging at 1.5 tesla (T) and at higher magnetic field MR systems. Human sized porcine models are routinely used to directly measure RF heating and correlate RF heating with the average Specific Absorption Rate (SAR). Novel MR thermometry methods and vasculature based validated bioheat thermal models are being developed for various organs to predict and optimize RF heating during imaging at the highest fields.

Research Interests:

• RF heating, its thrmophysiological consequences, and associated mechanisms
• Novel/Improved MR thermometry approaches to measure RF heating
• Bioheat transfer thermal modeling and Thermodynamics

In vivo thermoregulatory temperature responses of mammalian brains and thermo-susceptible human subjects (e.g., children, pregnant women, elderly) to RF heating are unknown for ultra-high field magnetic resonance (UHF-MR) systems. Studying these responses are necessary to develop appropriate RF safety guidelines for human MR imaging at the highest fields. More RF power is non-uniformly deposited in an imaged tissue with an increase in the field strength of an MR system for the same pulse sequence. Non-uniform RF heating (in vivo temperature distribution) results.

RF heating is a function of an RF coil, head geometry, tissue types, blood flow, and the physiological status of a biological model (i.e., un-anesthetized vs anesthetized, drugged vs un-drugged, fever or no fever, etc.). However, live humans can not be used to study the thermoregulatory temperature responses of RF heating at ultra-high fields. Therefore, animal models with thermoregulatory mechanisms similar to humans are needed to study physiological effects of a temperature change and temperature-time history. Perfused human cadavers are needed to determine RF heating. The results from the animal models and the humans cadavers are needed to be interpreted together to improve human RF safety at the highest fields. Dr. Shrivastava's work focuses on determining RF heating,its thermo-physiological consequences and associated mechanisms using animal models, cadavers, bioheat thermal modeling, and NMR thermometry to ensure RF safety for humans wearing/not wearing conductive medical implants.

More data and understanding of RF heating are desired for humans with implanted conductive medical devices, and imaged at 1.5 T and MR systems with higher field strenghts. Clinically harmful in vivo heating at the interface between an implant and tissue may be caused under certain, yet to be fully understood MR imaging conditions. The 1.5 T and 3 T MR systems are extensively used for clinical diagnostics and research in patients suspected of harboring various neurological disorders, including stroke and cancer. The 7 T system is increasingly used for human research world wide. Patients with conductive medical implants are as susceptible to neurological disorders, strokes and cancer as the general population. Therefore, it is desirable to be able to use MRI in the increasing population of patients with conductive implants. Further, signal to noise ratio (SNR), and spatial as well as temporal image resolution are directly proportional to the magnetic field strength in an MR system. Thus, with an increase in the field strength image resolution for the whole body improves dramatically. Improved resolution creates the potential for improved MRI-based targeting of implants, leading to greater surgical accuracy and postoperative efficacy. Thus, the availability of this versatile imaging tool to patients wearing implants will improve patient care, and basic and clinical research. Dr. Shrivastava's work focuses on developing clinically feasible protocols to improve targeting of and imaging around conductive implants with minimum RF heating. Porcine models and cadavers are used together with bioheat thermal modeling and NMR thermometry to develop the protocols.

Selected Publications: