A5100342775- Visual Attention and Saliency Detection
- Dementia and Cognitive Impairment Research
- Brain Tumor Detection and Classification
- Gaze Tracking and Assistive Technology
- Neurological Disease Mechanisms and Treatments
- Face and Expression Recognition
- Functional Brain Connectivity Studies
- Medical Image Segmentation Techniques
- Radiomics and Machine Learning in Medical Imaging
- Video Surveillance and Tracking Methods
- Medical Imaging Techniques and Applications
- Visual perception and processing mechanisms
- Advanced Neuroimaging Techniques and Applications
- Advanced Vision and Imaging
- Alzheimer's disease research and treatments
- Cerebrovascular and Carotid Artery Diseases
- Face recognition and analysis
- EEG and Brain-Computer Interfaces
- Advanced Image and Video Retrieval Techniques
- Advanced MRI Techniques and Applications
- Gene expression and cancer classification
- Collaboration in agile enterprises
- Face Recognition and Perception
- AI in cancer detection
- Fetal and Pediatric Neurological Disorders
Irvine Valley College
2024
Neurology, Inc
2020-2024
Korea Institute of Science & Technology Information
2012-2023
Korea University of Science and Technology
2023
Seoul St. Mary's Hospital
2021
Catholic University of Korea
2021
Yeouido St. Mary's Hospital
2021
LG (South Korea)
2020
Kyungpook National University
2006-2017
Sejong University
2015
Multi-label brain segmentation from magnetic resonance imaging (MRI) provides valuable structural information for most neurological analyses. Due to the complexity of algorithm, it could delay delivery neuroimaging findings. Therefore, we introduce Split-Attention U-Net (SAU-Net), a convolutional neural network with skip pathways and split-attention module that segments MRI scans. The proposed architecture employs blocks, pyramid levels, evolving normalization layers. For efficient training,...
Objective Alzheimer’s disease (AD) is the most common type of dementia and prevalence rapidly increased as elderly population worldwide. In contemporary model AD, it regarded a continuum involving preclinical stage to severe dementia. For accurate diagnosis monitoring, objective index reflecting structural change brain needed correctly assess patient’s severity neurodegeneration independent from clinical symptoms. The main aim this paper develop random forest (RF) algorithm-based prediction...
Alzheimer's disease (AD) accounts for 60-70% of the population with dementia. Mild cognitive impairment (MCI) is a diagnostic entity defined as an intermediate stage between subjective decline and dementia, about 10-15% people annually convert to AD. We aimed investigate most robust model modality combination by combining multi-modality image features based on demographic characteristics in six machine learning models. A total 196 subjects were enrolled from four hospitals Disease...
Brain segmentation of stroke patients can facilitate brain modeling for electrical non-invasive stimulation, a therapy stimulating function using an electric current. However, it remains challenging owing to its time-consuming, labor-dependent, and complicated pipeline. In addition, conventional tools that define lesions into one region rather than distinguishing between the stroke-affected regions cerebrospinal fluid lead inaccurate treatment results. this study, we first novel as detailed...
Background: The Fazekas scale is one of the most commonly used visual grading systems for white matter hyperintensity (WMH) brain disorders like dementia from T2-fluid attenuated inversion recovery magnetic resonance (MR) images (T2-FLAIRs). However, suffers low-intra and inter-rater reliability high labor-intensive work. Therefore, we developed a fully automated system using quantifiable measurements. Methods: Our approach involves four stages: (1) deep learning-based segmentation...
White-matter hyperintensity (WMH) is a primary biomarker for small-vessel cerebrovascular disease, Alzheimer’s disease (AD), and others. The association of WMH with brain structural changes has also recently been reported. Although fluid-attenuated inversion recovery (FLAIR) magnetic resonance imaging (MRI) provide valuable information about WMH, FLAIR does not other normal tissue information. multi-modal analysis T1-weighted (T1w) MRI thus desirable WMH-related aging studies. In clinical...
Normative brain magnetic resonance imaging (MRI) is essential to interpret the state of an individual’s health. However, a normative study often expensive for small research groups. Although several attempts have been made establish MRI norms, focus has limited certain age ranges. This aimed East Asian data using multi-site and determine robustness these clinical research. was gathered covering wide range cognitively normal populations (age: 18–96 years) from two open sources three sites....
Amyloid positron emission tomography (PET) scan is clinically essential for the non-invasive assessment of presence and spatial distribution amyloid-beta deposition in subjects with cognitive impairment suspected to have been a result Alzheimer's disease. Quantitative can enhance interpretation reliability PET scan; however, its clinical application has limited due complexity preprocessing. This study introduces novel deep-learning-based approach SUVR quantification that simplifies...
Inspired by the recent advances in generative models, we introduce a human action generation model order to generate consecutive sequence of motions formulate novel actions. We propose framework an autoencoder and adversarial network (GAN) produce multiple actions conditioned on initial state given class label. The proposed is trained end-to-end fashion, where jointly with GAN. NTU RGB+D dataset show that can different styles Moreover, successfully labels as conditions. conventional...
White matter hyperintensities (WMHs) are lesions in the white of brain that associated with cognitive decline and an increased risk dementia. The manual segmentation WMHs is highly time-consuming prone to intra- inter-variability. Therefore, automatic approaches gaining attention as a more efficient objective means detect monitor WMHs. In this study, we propose AQUA, deep learning model designed for fully from T2-FLAIR scans, which improves upon our previous study small lesion detection...
A scaler is one of the most important modules in various video applications, such as ultra-high definition TV and scalable systems. variety scaling techniques have been used to increase quality when resolution source image has be up- down-scaled. Some conventional schemes exploit property local block data. Others consider edge information data scaled. In this paper, we formulate a problem minimize loss resulting from resizing process. The considered both spatial frequency domains, then it...
We propose a new method for region of interest (ROI) based image segmentation that uses biologically motivated selective attention model. One the most important issues in on is how to decide upon semantic object according specific purpose. The proposed saliency map model conjunction with top-down Fuzzy adaptive resonance theory (ART) human interaction can generate scan path contains plausible area natural scene. In order extract an interesting generated by model, we extraction algorithm...
Imaging markers of cerebral small vessel disease provide valuable information on brain health, but their manual assessment is time-consuming and hampered by substantial intra- interrater variability. Automated rating may benefit biomedical research, as well clinical assessment, diagnostic reliability existing algorithms unknown. Here, we present the results \textit{VAscular Lesions DetectiOn Segmentation} (\textit{Where VALDO?}) challenge that was run a satellite event at international...
This study presents a novel semi-supervised deep learning method utilizing 3D TOF-MRA images for the detection of brain aneurysms, employing landmark-based techniques that mimic radiologist practices to enhance accuracy and efficiency.
Background: Application of visual scoring scales for regional atrophy in Alzheimer’s disease (AD) clinical settings is limited by their high time cost and low intra/inter-rater agreement. Objective: To provide automated using objective volume driven from deep-learning segmentation methods AD subtype classification magnetic resonance imaging (MRI). Methods: We enrolled 3,959 participants (1,732 cognitively normal [CN], 1594 with mild cognitive impairment [MCI], 633 AD). The occupancy indices...