Seong-Gi Kim, Ph.D.,is the Director of Neuroscience Imaging Research Center in the Institute for Basic Science (IBS), and Professor of Biomedical Engineering in Sungkyunkwan University (SKKU). He did his graduate works on in vivo NMR spectroscopy (1984-88) at Washington University, and postdoc training on structural biology at the University of Washington. In 1991, he moved to the Center for Magnetic Resonance Research in the University of Minnesota and involved into the first human fMRI studies in 1992. After advancing his rank to full Professor, he moved to the University of Pittsburgh at 2002.He was appointed as the inaugural Paul C. Lauterbur Chair in Imaging Research at 2009, which was created for honoring a Nobel laureate and MRI inventor. Recently he returned to Korea for setting up a new imaging center. His major research focus is to develop magnetic resonance imaging techniques for measuring brain physiology and function, to determine relationships between neural activity and hemodynamic responses, and to apply imaging tools for answering neuroscience questions.

Functional magnetic resonance imaging (fMRI) is widely usedfor human brain mapping due to its non-invasiveness and high spatial resolution, and complements to high-temporal-resolution EEG and MEG. Unlike EEG and MEG, fMRI measures neural activity indirectly via hemodynamic responses. Thus, it is critically important to understand the relationship between neural activity and fMRI responses. One important question is whether fMRI pinpoints active neural sites precisely even at a sub-millimeter scale. To investigate the accuracy of fMRI, well-established animal models can be used. Orientation-selective domain in a cat visual cortex can be selectively activated by presenting orthogonal visual stimuli (horizontal vs. vertical). Similarly layer-specific neural activity can be modulated. Therefore, we have investigated the spatial specificity of fMRI with animal column and layer model systems. We found thatBOLD fMRI is sensitive to draining vessels, and has poor spatial specificity at sub-millimeter resolution. By the contrary, cerebral blood volume-based fMRI is specific to active cortical fine structures, allowing us to map cortical columns and layers.

Dong Soo Lee, M.D.Ph.D.

1. Dept of Nuclear Medicine, Seoul National University Hospital, dsl@plaza.snu.ac.kr

2. Current appointments:

• Professor, Department of Nuclear Medicine and Professor and Chairman, Department of Molecular Medicine and Biopharmaceutical Sciences, Seoul National University
• Director, BioMAX/N-Bio Seoul National University
• Past-President of Korean Society of Human Brain Mapping, of Korean Society for Nanomedicine, and of Korean Society of Nuclear Medicine

• Published >300 articles in SCI journal (1999-now) including Nature, Nature Protocol.
• Editor-in-Chief,Nucl Med Mol Imaging (2014- )
• Editorial Board of Journal of Nuclear Medicine, European Journal of Nuclear Medicine and Molecular Imaging, Journal of Nuclear Cardiology, European Journal of Nanomedicine
  

Brain can be represented by graph. Structurally brain regions are connected directly or indirectly. Thresholding of correlation matrix will yield a brain graph and with lenient threshold it constructs an entirely connected graph. Functionally, brain works as a collection of seemingly fragmented clusters and this was the basis of the pursuit of the past two decades to understand brain connectivity. Beside graph theory or network science, another interpretation of brain connectivity came from algebraic topology based on the persistent homology. Simplicial chain complex was used to represent the 0, 1, 2-dimensional invariant of the hierarchical clustering of brain regions to perform or not to perform (resting state) taskson fMRI or other modalities. Boundary function of chain complexes depicts the image of the higher dimension and the kernel to the void in the lower dimension.
Using this framework of persistent homology of simplicial complex filtration (or Rips filtration in 1-dimension), one could successfully represent the invariant of Betti-0 (number of connected components) as barcodes (shuffled dendrogram), dendrogram or single linkage matrix and could find the differences thereof between normal children and autism on FDG PET, between prelingual deaf and postlingual deaf on T1 MRI, between auditory, visual and audio-visual sentence perception on activation fMRI, between normal and epilepsy model rats on FDG PET, or between normal and Parkinson disease model mice on FDG PET. Finally, this method was used for EEG data interpretation and for MEG analysis.
Taking an example of MEG data analysis, gating of memory encoding was re-analyzed with the data once reported for cue-alpha/ item-gamma power coupling and again reported for phase-power coupling. In our third analysis of the same data in the frame work of persistent homology, time-delayed cross-frequency source networks were found using power-power correlation between cue (alpha) and item (gamma) tasks of memory gating. Threshold-free representation of multi-regional correlation on persistent homology using negative correlation between all the brain regions yielded which regional pairs were associated with instruction compliance (gating of memory encoding) or memory performance.
I propose that the use of this method be extended to the analysis of causality between regional paired activities on brain graphs on MEG as well as other modalities. If this analysis picks up successfully the hub or controllable brain regions, we might intervene on this region to affect the function of this brain region using rTMS/ tCDS or transcranial ultrasound stimulation. The success of these intervention shall also be assessed again using the same method of connectivity in the framework of persistent homology.

Gen Uehara, Dr. Eng., is the Director of Applied Electronics Laboratory of Kanazawa Institute of Technology. He carried out research work on Josephson junction as a research associate at the University of Tokyo from 1981 to 1986. He joined Yokogawa Electric Corporation in 1986to develop SQUID magnetometers for biomagnetic applications. In 1990, he joined the National Project of Superconducting Sensor Laboratory to develop multi-channel SQUID MEG systems. Since 1996, was doing research and development of practical MEG systems as a visiting professor at Kanazawa Institute of Technology. Since 2007, he has been involved in education as a professor at Graduate School of Kanazawa Institute of Technology. Since he was promoted to be the Director of Applied Electronics Laboratory in 2010, he has been doing research for the expansion of MEG application and development of MSG, MCG and low field MRI.

One of the merits of MEG is its high temporal resolution in comparison with other brain diagnostic modalities such as fMRI and NIRS, because MEG directly measures neuronal signal, whereas fMRI and NIRS measure the metabolism in the blood of the brain. Another merit of MEG is its high spatial resolution in comparison with EEG, because the magnetic field emitted from the current in brain neurons is not distorted thanks to the homogeneous permeability of human head, whereas the voltage measurement with EEG suffers the distortion due to the inhomogeneous conductivity of the human head.We have developed supine mode MEG systems, which employ low Tc SQUID gradiometers. The subjects are to lie on the bed in a relaxed manner putting his/her head into the concave part of the liquid helium dewar. These systems are used for studies of the possibility of early stage detection of dementia, ischemia in the brain, autism spectrum disorder, and even for the research in linguistics.
The potential of MEG to localize the nerve activity non-invasively with high spatial/temporal resolution also attracts the interests of orthopaedists and neurologists who treat degenerative disorders of spinal cords, such as myelopathy. Neurological examination and observation of the morphological images acquired by MRI are considered insufficient for the localization of the spinal cord lesion because they often cause a lot of false-positive findings. Therefore, the functional test based on neurophysiology is indispensable for the accurate diagnosis of the spinal cord disorders. MSG, or magnetospinography, is one of the promising devices to realize non-invasive functional imaging of the spinal cord that allows medical doctors to avoid unnecessary surgery.

Dr. Qun Wang, professor in neurology, is the director of epilepsy division of neurology department, Beijing Tiantan Hospital, Capital Medical University as well as associate director of Beijing Key Laboratory of Translational Medicine for Cerebrovascular Disease. He is also a principal investigator in the Beijing Institute for Brain Disorders.
His research mainly focuses on clinical neurophysiology, neuroimaging and neuropharmacology. He reported the first case of depth electrode investigation and laser thermal therapy for insular seizure in 2013. He is working on a research project to investigate epileptic network monitoring by combining high density EEG to fMRI and DTI imaging technologies. In the past few years, he has explored the pharmacological mechanisms of polyphenol protection against neuronal damage-induced by seizure and cerebral ischemia/reperfusion. As the principal investigator, he has been funded with two research projects supported by NSFC (National Science Foundation of China), published more than 50 peer-reviewed papers and contributed 3 book chapters. He is an Ad hoc reviewer for Neuroscience Letters, Translational Neurosciences & Clinics， Pharmaceutical Biology，PLoS One, Journal of Neurological Disorders. He is a board member for Chinese Association against Epilepsy, Beijing Association against Epilepsy, Chinese Physician Association for Neurologists as well as a member for International Society for Neurochemistry, American Society for Neuroscience, American Epilepsy Society and American Stroke Association. He has worked as a postdoctoral fellow and research scientist at University of Missouri-Columbia (2002.06-2010.08) as well as clinical neurophysiologist at Washington University in St. Louis (2010.08-2012.12).

Background: MRI guided laser-induced thermal ablation for epilepsy is an exciting new minimally invasive technology with an emerging use for lesionectomy of a variety of epileptogenic focuses, especially some deep onset seizures such as hypothalamic hamartoma and insular seizure.
Objective: We utilized intraoperative MRI-guided stereotactic depth electrode recording and laser interstitial thermal therapy (LITT) for treatment of a patient with insular epileptic seizures.
Methods: A 54 year old man with a history of refractory complex partial seizures for 17 years underwent video-EEG, which showed seven complex partial seizures with left hemispheric diffuse EEG onset and evolution. A brain MRI revealed encephalomalacia in the left insula. A PET scan revealed hypometabolism in the left insula. An ictal SPECT injection during a typical complex partial seizure revealed hyperperfusion in the left insula in the region of MRI-defined encephalomalacia. He underwent stereotactic placement of left insular depth electrodes and left frontotemporal strip electrodes with intraoperative stereotactic MRI navigation. The intracranial video-EEG showed five complex partial seizures with left insular onset followed by left frontal and temporal propagation.
Ablation of the insular lesion using laser ablation interstitial thermal therapy (LITT) was performed to target the seizure-onset zone, and surrounding encephalomalacia, including both the anterior, and posterior insula.
Results: Postoperatively, the patient reported symptoms of anxiety, with associated mild dysarthria, which improved over the first six months of follow-up. He remained seizure free at his last follow-up, two years after surgery.
Conclusion: This case shows multimodality localization of left insular seizures, followed by LITT for surgical treatment of the epileptogenic focus, demonstrating usefulness of these diagnostic and treatment approaches for insular seizures. MRI-guided laser-induced thermal ablation for epilepsy is an exciting new minimally invasive technology and long-term results should be followed up and analyzed.