• Jong Chul Ye
  • Endowed Chair Professor

Curriculum Vitae


Ph.D. Electrical Engineering, Purdue University, USA (1999)
M.S. Control Engineering (currently Electrical Engineering), Seoul National University, Korea (1995)
B.S. Control Engineering (currently Electrical Engineering), Seoul National University, Korea (1993)

Professional Experience

KAIST, Department of Bio and Brain Engineering, Assistant, Associate, and Professor (tenured) (2004 - current)
KAIST, Department of Electrical Engineering, Adjunct Professor (2007 - current)
GE Global Research Center(Niskayuna, New York), X-ray CT Technology Group, Senior Researcher (2003 - 2004)
Philips Research Center (Briarcliff Manor, New York), Senior Member Research Staff (2001 - 2003)

Research Interests

Biomedical Imaging, Machine Learning

Honors & Awards

March 216: KAIST Endowed Chair Professorship
Feb. 2012 - Jan. 2013 : Beckman Senior Fellowship Award, Univ. of Illinois at Urbana-Champaign
Feb. 2012 : KAIST Research Excellence Award



Deep learning for biomedical image reconstruction

Recently, deep learning approaches with various network architectures have achieved significant performance improvement over existing iterative reconstruction methods in various imaging problems. However, it is still unclear why these deep learning architectures work for specific inverse problems. To address these issues, here we show that the long-searched-for missing link is the convolutional framelets for representing a signal by convolving local and non-local bases. This discovery reveals the limitations of many existing deep learning architectures for inverse problems, and leads us to propose a novel deep convolutional framelets} neural network. Using extensive experiments with x-ray computed tomography (CT), magnetic resonance imaging (MRI), microscopy, diffuse optical tomography, ultrasound, etc, we demonstrated that our deep convolution framelets network shows consistent improvement over existing reconstruction methods.