Speaker
Description
Deep neural network techniques have improved various signal processing results in many areas. Medical imaging field is a representative area affected by the deep learning technology. MRI system and image reconstruction methods have been studied in KAIST since 1980. This talk briefly reviews the history of KAIST MRI research and introduces current progresses in MRI. Two-tesla MRI system, which was one of the highest field MRI system in the beginning of 1980's, had been developed in KAIST in 1985. Using this system, various imaging techniques had been developed and presented.
Recently, methods for quantification of diffusion parameters and chemical exchange saturation transfer (CEST) parameters have been developed. And a new motion correction method was developed. All of these methods utilized deep neural networks. This talk introduces the latest progress in MRI research at KAIST.
References
W.I. Lee, B.J. Kim, and H.W. Park, “Quantification of intravoxel incoherent motion with optimized b-values using deep neural network,” Magnetic Resonance in Medicine, Vol. 86(1), pp. 230-244, July 2021.
J.Y. Lee, B.J. Kim, and H.W. Park, “MC2-net: motion correction network for multi-contrast brain MRI,” Magnetic Resonance in Medicine, Vol. 86(2), pp. 1077-1092, Aug. 2021.
B.G. Kang, B.J. Kim, H.W. Park, Almonak, and H.Y. Heo, “Learning-based optimization of acquisition schedule for magnetization transfer contrast MR fingerprinting,” NMR in Biomedicine, Vol. 35, e4662, 2022.
| Keywords | MRI, image reconstruction, Deep learning |
|---|
