A5052550275- Radiomics and Machine Learning in Medical Imaging
- MRI in cancer diagnosis
- Liver Disease Diagnosis and Treatment
- Nutrition and Health in Aging
- Hepatocellular Carcinoma Treatment and Prognosis
- Liver Disease and Transplantation
- Body Composition Measurement Techniques
- Advanced MRI Techniques and Applications
- Advanced X-ray and CT Imaging
- Advanced Neuroimaging Techniques and Applications
- Organ Transplantation Techniques and Outcomes
- Neuroscience and Neuropharmacology Research
- Lung Cancer Diagnosis and Treatment
- Lanthanide and Transition Metal Complexes
- Frailty in Older Adults
- Agriculture, Soil, Plant Science
- Toxic Organic Pollutants Impact
- Meningioma and schwannoma management
- Neurofibromatosis and Schwannoma Cases
- Smart Agriculture and AI
- Renal cell carcinoma treatment
- Drug Transport and Resistance Mechanisms
- Retinal and Optic Conditions
- Brain Tumor Detection and Classification
- Molecular Sensors and Ion Detection
University of Ulsan
2015-2024
Asan Medical Center
2015-2024
Ulsan College
2015-2024
National Chung Hsing University
2022-2023
Research Institute of Radiology
2014-2019
Athinoula A. Martinos Center for Biomedical Imaging
2017
Harvard University
2017
Massachusetts General Hospital
2017
Creative Commons
2016
National Cancer Center
2016
Purpose To develop and validate a deep learning system (DLS) for staging liver fibrosis by using CT images in the liver. Materials Methods DLS CT-based of was created development data set that included portal venous phase 7461 patients with pathologically confirmed fibrosis. The diagnostic performance evaluated separate test sets 891 patients. influence patient characteristics techniques on accuracy logistic regression analysis. In subset 421 patients, compared radiologist's assessment,...
Purpose To develop and validate a radiomics-based model for staging liver fibrosis by using gadoxetic acid–enhanced hepatobiliary phase MRI. Materials Methods In this retrospective study, 436 patients (mean age, 51 years; age range, 18–86 319 men [mean years]; 117 women 50 18–79 years]) with pathologic analysis–proven who underwent MRI from June 2015 to December 2016 were randomized in three-to-one ratio into development (n = 329) test 107) cohorts, respectively. the cohort, was developed...
Abstract Background The impact of sarcopenia on clinical outcomes coronavirus disease 2019 (COVID-19) is not clearly determined yet. We aimed to investigate the association between baseline and in patients with COVID-19. Methods All hospitalized adult COVID-19 who had chest computed tomography (CT) scans at a Korean university hospital from February 2020 May were included. main outcome was time admission discharge. Death considered as competing risk for Baseline skeletal muscle...
Quantification of abdominal muscle mass by cross-sectional imaging has been increasingly used to diagnose sarcopenia; however, the technical method for quantification not standardized yet. We aimed determine an optimal measure area.Among 50 consecutive subjects who underwent CT and MRI possible liver donation, total area (TAMA) psoas (TPA) at L3 inferior endplate level were measured two blinded readers. Inter-scan agreement between inter-reader readers evaluated using intraclass correlation...
Measurement of the liver and spleen volumes has clinical implications. Although computed tomography (CT) volumetry is considered to be most reliable noninvasive method for volume measurement, it limited application in practice due its time-consuming segmentation process. We aimed develop validate a deep learning algorithm (DLA) fully automated using portal venous phase CT images various conditions.A DLA was trained development dataset from 813 patients. Performance evaluated two separate...
Abstract As sarcopenia research has been gaining emphasis, the need for quantification of abdominal muscle on computed tomography (CT) is increasing. Thus, a fully automated system to select L3 slice and segment in an end-to-end manner demanded. We aimed develop deep learning model (DLM) with consideration anatomic variations cross-sectional areas (CSAs) fat. Our DLM, named L3SEG-net, was composed YOLOv3-based algorithm selecting convolutional network (FCN)-based segmentation. The developed...
Single-plant growth monitoring aids precision agricultural decision-making to reduce the costs related pesticides, fertilizers, and labor. This study integrated visible/multi-spectral UAV imagery with two deep learning methods, object detection semantic segmentation, obtain a visualized map that could assist in precise field management for broccoli cultivation. For plant detection, feature extraction was conducted using multiscale dilated convolution, which enabled effective of images taken...
The purpose of this article is to assess the value computer-aided diagnosis (CAD) for prostate cancer detection on dynamic contrast-enhanced MRI (DCE-MRI).DCE-MRI examinations 42 patients with were used generate perfusion parameters, including baseline and peak signal intensities, initial slope, maximum slope within 50 seconds after contrast injection (slope(50)), wash-in rate, washout time peak, percentage relative enhancement, enhancement ratio, arrival, efflux rate constant from...
Purpose To evaluate the influence of fitting methods on accuracy and reliability intravoxel incoherent motion (IVIM) parameters, with a particular emphasis constraint function. Materials Methods Diffusion‐weighted (DW) imaging data were analyzed using IVIM‐based full‐fitting (simultaneous fit all parameters) segmented‐fitting (step‐by‐step each parameter), without function, to estimate molecular diffusion coefficient ( D slow ), perfusion fraction f flow‐related fast ). Computational...
In this paper, we propose a novel method of identifying pulmonary nodules in lung CT. Specifically, devise deep neural network by which extract abstract information inherent raw hand-crafted imaging features. We then combine the learned representations with original features into long feature vector. By taking combined vectors, train classifier, preceded selection via t-test. To validate effectiveness proposed method, performed experiments on our in-house dataset 20 subjects; 3,598...
To qualitatively and quantitatively compare the diagnostic performance of rs-EPI (readout segmented echo planar imaging) reduced FOV (field-of-view) EPI in patients with biopsy-proven breast cancer at 3T.Between November 2013 July 2014, 96 (age range, 30-75 years: mean, 52 years) were retrospectively enrolled this study. In all patients, rFOV performed using a 3T MR scanner. Differences between two sequences compared by measuring tumor apparent diffusion coefficient (ADC), signal-to-noise...
Purpose To evaluate the diagnostic value of apparent diffusion coefficient (ADC) and intravoxel incoherent motion (IVIM) parameters in differentiating patients with either a normal pancreas (NP), pancreatic ductal adenocarcinoma (PDAC), neuroendocrine tumor (NET), solid pseudopapillary (SPT), acute pancreatitis (AcP), vs. autoimmune (AIP). Materials Methods In all, 84 pathologically confirmed tumors (60 PDACs, 15 NETs, 9 SPTs), 20 (13 AcPs, 7 AIPs), 30 NP subjects underwent IVIM...
Muscle quality is associated with fatty degeneration or infiltration of the muscle, which may be decreased muscle function and increased disability.The aim this study to evaluate feasibility automated quantitative measurements skeletal on computed tomography (CT) images assess normal-attenuation myosteatosis.We developed a web-based toolkit generate map by categorizing components. First, automatic segmentation total abdominal area (TAMA), visceral fat area, subcutaneous was performed using...
The increased cochlear signal on FLAIR images in patients with acoustic neuroma is explained by an concentration of protein the perilymphatic space. However, there still debate whether a correlation between and degree hearing disturbance neuroma. Our aim was to investigate clinical significance 3D according extent large patient cohort.This retrospective study enrolled 102 neuroma, who were divided into 2 groups based tumor location; 22 tumors confined internal auditory canal 80 extended...
To investigate the correlation between perfusion- and diffusion-related parameters from intravoxel incoherent motion (IVIM) those dynamic contrast-enhanced MR imaging (DCE-MRI) diffusion-weighted in tumors normal muscles of head neck.We retrospectively enrolled 20 consecutive patients with neck performed using a 3T scanner. Tissue diffusivity (D), pseudo-diffusion coefficient (D(*)), perfusion fraction (f) were derived bi-exponential fitting IVIM data obtained 14 different b-values three...
We aimed to develop and test a deep learning algorithm (DLA) for fully automated measurement of the volume signal intensity (SI) liver spleen using gadoxetic acid-enhanced hepatobiliary phase (HBP)-magnetic resonance imaging (MRI) evaluate clinical utility DLA-assisted assessment functional capacity.The DLA was developed HBP-MRI data from 1014 patients. Using an independent dataset (110 internal 90 external MRI data), segmentation performance measured Dice similarity score (DSS), agreement...