Deep learning for Parallel Transmission
Parallel RF transmission (pTx) is an essential technique for high-field MRI in which different RF waveforms are simultaneously played through the elements of a multi-coil transmit array, to shape the combined transmit RF field dynamically.
We are working on the ‘smart’ pTx using deep learning methods on massive human models to predicate a universal device and pulse for a large population without the need to adjust the pulse for every subject. We are also working on pTx to shape the magnetic and electric fields to reduce the RF heating near the implants.
These projects focus on algorithm development and electromagnetic simulation. They are well-suited for students with a strong background in physics or mathematics.
More details can be found in our publications:
https://onlinelibrary.wiley.com/doi/full/10.1002/mrm.26020
https://www.sciencedirect.com/science/article/pii/S0730725X22001205