Global-Leading Research and Education in Brain and Cognitive engineering KOREA UNIVERSITY To understand the structure and the key principles of high-order information processing by which brains work,
and to implement these in artificial systems that interact intelligently with real world

Research Highlight

Research
Prof.Heung-II Suk’s paper accepted in Medical Image Analysis
Author Administrator (IP: *.152.74.116) Date 2017-01-13 10:40 Views 458
원본사진보기

Prof. Heung-II Suk’s new research on `Deep Ensemble Learning of Sparse Regression Models for Brain Disease Diagnosis` was accepted in Medical Image Analysis.

 

Title: Deep Ensemble Learning of Sparse Regression Models for Brain Disease Diagnosis

Abstract: 

Recent studies on brain imaging analysis witnessed the core roles of machine learning techniques in computer-assisted intervention for brain disease diagnosis. Of various machine-learning techniques, sparse regression models have proved their effectiveness in handling high-dimensional data but with a small number of training samples, especially in medical problems. In the meantime, deep learning methods have been making great successes by outperforming the state-of-the-art performances in various applications. In this paper, we propose a novel framework that combines the two conceptually different methods of sparse regression and deep learning for Alzheimer's disease/mild cognitive impairment diagnosis and prognosis. Specifically, we first train multiple sparse regression models, each of which is trained with different values of a regularization control parameter. Thus, our multiple sparse regression models potentially select different feature subsets from the original feature set; thereby they have different powers to predict the response values, i.e., clinical label and clinical scores in our work. By regarding the response values from our sparse regression models as target-level representations, we then build a deep convolutional neural network for clinical decision making, which thus we call `Deep Ensemble Sparse Regression Network.' To our best knowledge, this is the first work that combines sparse regression models with deep neural network. In our experiments with the ADNI cohort, we validated the effectiveness of the proposed method by achieving the highest diagnostic accuracies in three classification tasks. We also rigorously analyzed our results and compared with the previous studies on the ADNI cohort in the literature.

File 57feda4997e7d.gif(10.9K)


Research
Prof. Seong-Whan Lee's paper accepted in Pattern Recognition
Prof. Seong-Whan Lee's paper accepted in Pattern Recognition
Prof. Seong-Whan Lee's new research,  Discriminative Context Learning with Gated Recurrent Unit for Group Activity Recognition was accepted in Pattern Recognition.  Title: Discriminative Context Learning with Gated Recurrent Unit for Group Activity Recognition  Abstract In this study, we address the problem of similar local motions that create confusion within different group activities. To reduce the influences…

2017-11-13

Prof. Byoung-Kyong Min's paper accepted in Hearing Research
Prof. Byoung-Kyong Min's paper accepted in Hearing Research
Prof. Byoung-Kyong Min's new research, The Absence of Resting-state High-Gamma Cross-Frequency Coupling in Patients with Tinnitus was accepted in Hearing Research.Title: The Absence of Resting-state High-Gamma Cross-Frequency Coupling in Patients with Tinnitus Abstract:Tinnitus is a psychoacoustic phantom perception of currently unknown neuropathology. Despite a growing number of post-stimulus tinnitus studies, uncertainty still exists regar…

2017-10-23

Prof. Dong-Joo Kim's paper accepted in Journal of Neurosurgical Anesthesiology
Prof. Dong-Joo Kim's paper accepted in Journal of Neurosurgical Anesthesiology
Prof. Dong-Joo Kim's new research,  Hemodynamic Instability and Cardiovascular Events After Traumatic Brain Injury Predict Outcome After Artifact Removal With Deep Belief Network Analysis was accepted in Journal of Neurosurgical Anesthesiology.Title: Hemodynamic Instability and Cardiovascular Events After Traumatic Brain Injury Predict Outcome After Artifact Removal With Deep Belief Network AnalysisAbstract:- Background: Hemodynamic ins…

2017-09-27

Prof. Christian Wallraven’s paper accepted in Brain Structure and Function
Prof. Christian Wallraven’s paper accepted in Brain Structure and Function
Prof. Christian Wallraven’s new research on Neuroanatomical Correlates of Haptic Object Processing: Combined Evidence from Tractography and Functional Neuroimaging was accepted in Brain Structure and Function.Title: Neuroanatomical Correlates of Haptic Object Processing: Combined Evidence from Tractography and Functional NeuroimagingAbstract: Touch delivers a wealth of information already from birth, helping infants to acquire knowledge ab…

2017-09-06

Prof. Seong-Whan Lee's paper accepted in Human Brain Mapping
Prof. Seong-Whan Lee's paper accepted in Human Brain Mapping
Prof. Seong-Whan Lee's new research, Multiple Functional Networks Modeling for Autism Spectrum Disorder Diagnosis was accepted in Human Brain Mapping. Title: Multiple Functional Networks Modeling for Autism Spectrum Disorder Diagnosis AbstractDespite countless studies on Autism Spectrum Disorder (ASD), diagnosis relies on specific behavioral criteria and neuroimaging biomarkers for the disorder are still relatively scarce and …

2017-08-08

Prof. Seong-Whan Lee's paper accepted in IEEE Trans. on Neural Systems and Rehabilitation Engineering
Prof. Seong-Whan Lee's paper accepted in IEEE Trans. on Neural Systems and Rehab…
Prof. Seong-Whan Lee's new research, Motion-based Rapid Serial Visual Presentation for Gaze-Independent Brain-Computer Interfaces was accepted in IEEE Trans. on Neural Systems and Rehabilitation Engineering. Title: Motion-based Rapid Serial Visual Presentation for Gaze-Independent Brain-Computer Interfaces  Abstract: Most event-related potential (ERP)-based brain-computer interface (BCI) spellers primarily use matrix layouts, …

2017-08-02

Prof. Dong-Joo Kim's paper accepted in Neurosurgery
Prof. Dong-Joo Kim's paper accepted in Neurosurgery
Prof. Dong-Joo Kim's new research, Abilities of a Densitometric Analysis of Computed Tomography Images and Hemorrhagic Parameters to Predict Outcome Favorability in Patients with Intracerebral Hemorrhagewas accepted in Neurosurgery. Title: Abilities of a Densitometric Analysis of Computed Tomography Images and Hemorrhagic Parameters to Predict Outcome Favorability in Patients with Intracerebral Hemorrhage Abstract: - Background: In…

2017-06-20

Prof. Byoung-Kyong Min's paper accepted in IEEE Transactions on Information Forensics and Security
Prof. Byoung-Kyong Min's paper accepted in IEEE Transactions on Information Fore…
Prof. Byoung-Kyong Min's new research, Individual Identification using Cognitive Electroencephalographic Neurodynamics was accepted in IEEE Transactions on Information Forensics and Security. Title: Individual Identification using Cognitive Electroencephalographic Neurodynamics Abstract:As the brain is a unique biological system that reflects the subtle distinctions in the mental attributes of individual humans, electroencephalogra…

2017-05-04

Prof. Byoung-Kyong Min's paper accepted in Trends in Biotechnology
Prof. Byoung-Kyong Min's paper accepted in Trends in Biotechnology
Prof. Byoung-Kyong Min's new research, Harnessing Prefrontal Cognitive Signals for Brain-Machine Interfaces was accepted in Trends in Biotechnology. Title: Harnessing Prefrontal Cognitive Signals for Brain-Machine Interfaces Abstract:Brain-machine interfaces (BMIs) enable humans to interact with devices by modulating their brain signals. Despite impressive technological advancements, several obstacles remain. The most commonly…

2017-03-15

Prof. Seong-Whan Lee's paper accepted in IEEE Trans. on Pattern Analysis and Machine Intelligence
Prof. Seong-Whan Lee's paper accepted in IEEE Trans. on Pattern Analysis and Mac…
Prof. Seong-Whan Lee's new research, A Novel Linelet-based Representation for Line Segment Detection was accepted in IEEE Trans. on Pattern Analysis and Machine Intelligence. Title: A Novel Linelet-based Representation for Line Segment Detection Abstract: This paper proposes a method for line segment detection in digital images.We propose a novel linelet-based representation to model intrinsic properties of line segments in rasterized image …

2017-02-14

 1 2 3 4 5 6 7 8 9