A5028390126- Radiomics and Machine Learning in Medical Imaging
- Neural dynamics and brain function
- Advanced Neuroimaging Techniques and Applications
- Functional Brain Connectivity Studies
- Brain Tumor Detection and Classification
- Medical Imaging Techniques and Applications
- Intracerebral and Subarachnoid Hemorrhage Research
- MRI in cancer diagnosis
- Parkinson's Disease Mechanisms and Treatments
- Neurological disorders and treatments
- Lung Cancer Diagnosis and Treatment
- Medical Image Segmentation Techniques
- Advanced Neural Network Applications
- Domain Adaptation and Few-Shot Learning
- EEG and Brain-Computer Interfaces
- Neuroscience and Neural Engineering
- Trigeminal Neuralgia and Treatments
- Diversity and Career in Medicine
- Neural and Behavioral Psychology Studies
- COVID-19 diagnosis using AI
- Healthcare professionals’ stress and burnout
- Cerebral Venous Sinus Thrombosis
- Advanced Memory and Neural Computing
- Migraine and Headache Studies
- Molecular Biology Techniques and Applications
Sungkyunkwan University
2020-2024
Institute for Basic Science
2020-2024
Ripon College
2018
University of Nebraska Medical Center
2018
Abstract Human nonverbal communication tools are very ambiguous and difficult to transfer machines or artificial intelligence (AI). If the AI understands mental state behind a user’s decision, it can learn more appropriate decisions even in unclear situations. We introduce Brain–AI Closed-Loop System (BACLoS), wireless interaction platform that enables human brain wave analysis transfers results verify enhance decision-making. developed earbud-like electroencephalography (EEG) measurement...
Abstract Classification of headache disorders is dependent on a subjective self-report from patients and its interpretation by physicians. We aimed to apply objective data-driven machine learning approaches analyze patient-reported symptoms test the feasibility automated classification disorders. The data 2162 were analyzed. Headache merged into five major entities. divided training (n = 1286) 876) cohorts. trained stacked classifier model with four layers XGBoost classifiers. first layer...
Deep learning (DL) is a breakthrough technology for medical imaging with high sample size requirements and interpretability issues. Using pretrained DL model through radiomics-guided approach, we propose methodology stratifying the prognosis of lung adenocarcinomas based on pretreatment CT. Our approach allows us to apply smaller enhanced interpretability. Baseline radiomics models were developed tested using local (n = 617) cohort. The further in an external validation 70) cohort was...
Breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is a sensitive technique critical for breast cancer diagnosis. However, the administration of contrast agents poses potential risk. This can be avoided if MRI obtained without using agents. Thus, we aimed to generate T1-weighted (ceT1) images from pre-contrast T1 weighted (preT1) in breast.
Functional hierarchy establishes core axes of the brain, and overweight individuals show alterations in networks anchored on these axes, particularly those involved sensory cognitive control systems. However, quantitative assessments hierarchical brain organization are lacking. Capitalizing stepwise functional connectivity analysis, we assess altered relative to healthy weight controls along hierarchy. Seeding from regions associated with obesity phenotypes, conduct analysis at different...
Distinguishing the autism spectrum disorder (ASD) from typical control (TC) using resting-state functional magnetic resonance imaging (rs-fMRI) is very difficult because ASD has heterogenetic properties and induce small changes in brain structure. Moreover, distinguishing TC data obtained many sites even more factors might negatively affect classification model leading to unstable results. This difficulty especially true for existing rs-fMRI analysis methods such as connectivity analysis....
Abstract Background Many studies have successfully identified radiomics features reflecting macroscale tumor and microenvironment for various organs. There is an increased interest in applying these found a given organ to other Here, we explored whether common could be over target organs vastly different environments. Methods Four datasets of three were analyzed. One model was constructed from the training set (lungs, n = 401), further evaluated independent test sets spanning 59; kidneys,...
Background and aim: Tumor staging in non-small cell lung cancer (NSCLC) is important for treatment prognosis. Staging involves expert interpretation of imaging, which we aim to automate with deep learning (DL). We proposed a cascaded DL method comprised two steps classification between early- advanced-stage NSCLC using pretreatment computed tomography. Methods: developed tested model classify training (n = 90), validation 8), test 37, n 26) cohorts obtained from the public domain. The first...
The "motor reserve" is an emerging concept based on the discrepancy between severity of parkinsonism and dopaminergic degeneration; however, related brain structures have not yet been elucidated.We investigated relevant to motor reserve in Parkinson's disease (PD) this study.Patients with drug-naïve, early PD were enrolled, who then underwent dopamine transporter (DAT) scan diffusion tensor imaging (DTI). symptoms was evaluated Unified Disease Rating Scale score bradykinesia rigidity more...
Imaging genetics is a methodology for discovering associations between imaging and genetic variables. Many studies adopted sparse models such as canonical correlation analysis (SCCA) genetics. These methods are limited to modeling the linear relationship cannot capture non-linear high-level explored Deep learning approaches under in genetics, compared their great successes many other biomedical domains image segmentation disease classification. In this work, we proposed deep model select...
Waiting impulsivity in progressive supranuclear palsy-Richardson's syndrome (PSP-RS) is difficult to assess, and its regulation known involve nucleus accumbens (NAc) subregions. We investigated waiting using the "jumping gun" (JTG) sign, which defined as premature initiation of clapping before start signal three-clap test compared clinical features PSP-RS patients with without sign analyzed neural connectivity microstructural changes NAc subregions.A positive JTG was participant starting...
Few national studies have examined the influence of role models as a potential predictor for caring medically underserved (MUS) patients. This study tested associations between previous physician model exposure and MUS populations, well examines practice environments these physicians.Between October December 2011, we mailed confidential questionnaire to representative sample 2000 US physicians from various specialties. The primary criterion variable was "Is your patient population considered...
The prognosis, hence survival, of patients with brain tumors is highly dependent on the size and grade tumor. Thus, joint learning tumor segmentation overall survival can benefit each other. In this work, we explored feasibility prediction patients' through U-Net guided by information segmentation. We evaluated proposed model multimodal (BraTS) 2017 challenge dataset. achieved mean Dice score 0.595 for test set. Pearson correlation coefficient set was 0.243, indicating promising results both...