Dr. Jong-Hwan Lee received his M.S. and Ph.D. degrees in Electrical Engineering and Computer Science from the Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, in 2000 and 2005, respectively. He was a research fellow from 2005 to 2008 at the Brigham and Women’s Hospital, Harvard Medical School, Boston, MA and then promoted to an instructor in 2008. Since fall 2009, he has been a faculty member at the Department of Brain and Cognitive Engineering, Korea University and currently an Associate Professor. His longstanding research interests have been in the investigation of hidden information underlying sensory signals and development of efficient methods for accurate analysis of this information. His recent research has focused on the application of the artificial neural networks and/or machine learning algorithms to the neuroimaging data, such as structural MRI (sMRI), functional MRI (fMRI), and electroencephalography (EEG). He has been working on the studies/projects such as (i) a novel application of independent vector analysis and (ii) an iterative approach of dual-regression with sparse constraints on spatial patterns of neuronal activations applying to group fMRI data processing, (iii) automated classification of thought processes measured via fMRI data, (iv) analysis of simultaneous EEG-fMRI methods and their applications to sleep-onset/epileptic-foci detection and to denoising of cryogenic pump artifact of MRI scanner, (v) real-time fMRI based neurofeedback, (vi) smoking cigarette addiction research via neuroimaging intervention, and (vii) an early diagnosis of neurodegenerative and/or neuropsychiatric diseases from sMRI and/or fMRI using advanced machine learning algorithms including deep learning using deep neural networks.

Lab homepage: http://bspl.korea.ac.kr

Education

• 03/2000-02/2005       PH.D.(Electrical Engineering and Computer Science), KAIST
• 03/1998-02/2000       M.S.(Electrical Engineering and Computer Science), KAIST
• 03/1994-02/1998       B.S. (Electronics Engineering), Yonsei University, Korea

Professional Experiences

• 04/2000 –08/2000
• Visiting Scholar, Institute for Neural Computation, University of California at San Diego, CA, USA
• 03/2005 – 06/2006
• Postdoctoral Researcher, KAIST, Daejeon, Korea
• 10/2005 – 05/2008
• Research Fellow, Dept. of Radiology, Brigham and Women's Hospital, Boston, MA
• Research Fellow, Dept. of  Radiology, Harvard Medical School, Boston, MA
• 06/2008 – 08/2009
• Research Associate, Dept. of Radiology, Brigham and Women's Hospital, Boston, MA
• Instructor, Dept. of Radiology, Harvard Medical School, Boston, MA
• 09/2009 – 08/2015
• Assistant Professor, Dept. of Brain and Cognitive Engineering, Korea University, Seoul
• 09/2016 - 08/2017
• Visiting faculty, Section on Functional Imaging Methods, Lab on Brain and Cognition, National Institute of Mental Health, National Institute of Health
• 09/2015 – Present
• Associate Professor, Dept. of Brain and Cognitive Engineering, Korea University, Seoul

Representative publications

• Jang H, Plis SM, Calhoun VD, Lee JH. Task-specific feature extraction and classification of fMRI volumes using a deep neural network initialized with a deep belief network: Evaluation using sensorimotor tasks. Neuroimage. 2017 Jan 15;145(Pt B):314-328. doi:10.1016/j.neuroimage.2016.04.003. Epub 2016 Apr 11.


• Kim J, Calhoun VD, Shim E, Lee JH, Deep neural network with weight sparsity control and pre-training extracts hierarchical features and enhances classification performance: Evidence from whole-brain resting-state functional connectivity patterns of schizophrenia, Neuroimage. 2016 Jan1;124(Pt A):127-46. doi: 10.1016/j.neuroimage.2015.05.018. Epub 2015 May 15.


• Kim DY, Yoo SS, Tegethoff M, Meinlschimidt G, Lee JH, The inclusion of functional connectivity information into fMRI-based neurofeedback improves its efficacy in the reduction of cigarette cravings, Journal of Cognitive Neuroscience, 2015 Aug;27(8):1552-72. doi: 10.1162/jocn_a_00802. Epub2015 Mar 11


• Kim HC, Yoo SS, Lee JH, Recursive approach of EEG segment based principal component analysis substantially reduces helium-pump artifacts of EEG data simultaneously acquired with fMRI, NeuroImage, 104:437-51, 2015;10.1016/j.neuroimage.2014.09.049


• Kim YH, Kim J. Lee JH, Iterative approach of dual regression with a sparse prior enhances the performance of independent component analysis for group functional magnetic resonance imaging (fMRI) data, NeuroImage, 2012; 63(4): 1864-1889.


• Lee JH, Marzelli M, Jolesz FA, Yoo SS, Automated classification of fMRI data employing trial-based imagery tasks. Medical Image Analysis. 2009; 13(3): 392-404.


• Lee JH, Lee TW, Jolesz FA, Yoo SS, Independent Vector Analysis (IVA): Multivariate Approach for fMRI Group Study. NeuroImage. 2008;40(1): 86-109.


• Lee JH, O’Leary HM, Park H, Jolesz FA, Yoo SS, Atlas-based Multi-channel Monitoring of Functional MRI Signals in Real-time: Automated Approach. Human Brain Mapping. 2008; 29 (2): 157-166.

Brain Science and Engineering using Neuroimaging Modality

Investigating brain functions measured by various neuroimaging modalities such as functional MRI (fMRI) and electroencephalography (EEG) toward potential applications such as brain healthcare systems including brain-computer interface (BCI) and brain-machine interface (BMI) to enhance human performance.

Now, the Question is,
Brain engineering and healthcare applications as well as enhancement of human performance via neuroimaging modalities are possible?

Aims

Our goal is to investigate brain functions measured via various neuroimaging modalities including functional MRI (fMRI) and electroencephalography (EEG) employing various signal processing techniques. We have done some interesting works including the fMRI data analyses using novel analytical methods such as independent vector analysis (IVA), iterative dual-regression of group independent component analysis (ICA) with a sparse prior to better estimate true neuronal activity, recursive principal component analysis (PCA) to EEG-segments of simultaneous EEG-fMRI data, and deep neural networks (DNN) to fMRI data. The developed methods would gainfully be applied to the neuroimaging data including fMRI, simultaneous EEG-fMRI, and real-time fMRI based neurofeedback method. Based on correct understanding of human brain functions, we would like to focus on the brain engineering including the BCI/BMI and ultimately on preclinical applications to develop an option to diagnose and treat the various neuropsychiatric illnesses such as depression, schizophrenia, and substance abuse. We believe that the proper analytical methods to exploit the hidden information of the neuroimaging data would lead to better understanding of the human brain and to better engineer the brain and ultimately toward enhancement of quality of life.