Deep Learning-Based Assessment of Functional Liver Capacity Using Gadoxetic Acid-Enhanced Hepatobiliary Phase MRI

Park Hyo Jung,Yoon Jee Seok,Lee Seung Soo,Suk Heung-Il,Park Bumwoo,Sung Yu Sub,Hong Seung Baek,류화성 대한영상의학회 2022 Korean Journal of Radiology Vol.23 No.7

Objective: We aimed to develop and test a deep learning algorithm (DLA) for fully automated measurement of the volume and signal intensity (SI) of the liver and spleen using gadoxetic acid-enhanced hepatobiliary phase (HBP)-magnetic resonance imaging (MRI) and to evaluate the clinical utility of DLA-assisted assessment of functional liver capacity. Materials and Methods: The DLA was developed using HBP-MRI data from 1014 patients. Using an independent test dataset (110 internal and 90 external MRI data), the segmentation performance of the DLA was measured using the Dice similarity score (DSS), and the agreement between the DLA and the ground truth for the volume and SI measurements was assessed with a Bland-Altman 95% limit of agreement (LOA). In 276 separate patients (male:female, 191:85; mean age ± standard deviation, 40 ± 15 years) who underwent hepatic resection, we evaluated the correlations between various DLA-based MRI indices, including liver volume normalized by body surface area (LVBSA), liver-to-spleen SI ratio (LSSR), MRI parameter-adjusted LSSR (aLSSR), LSSR x LVBSA, and aLSSR x LVBSA, and the indocyanine green retention rate at 15 minutes (ICG-R15), and determined the diagnostic performance of the DLA-based MRI indices to detect ICG-R15 ≥ 20%. Results: In the test dataset, the mean DSS was 0.977 for liver segmentation and 0.946 for spleen segmentation. The Bland- Altman 95% LOAs were 0.08% ± 3.70% for the liver volume, 0.20% ± 7.89% for the spleen volume, -0.02% ± 1.28% for the liver SI, and -0.01% ± 1.70% for the spleen SI. Among DLA-based MRI indices, aLSSR x LVBSA showed the strongest correlation with ICG-R15 (r = -0.54, p < 0.001), with area under receiver operating characteristic curve of 0.932 (95% confidence interval, 0.895–0.959) to diagnose ICG-R15 ≥ 20%. Conclusion: Our DLA can accurately measure the volume and SI of the liver and spleen and may be useful for assessing functional liver capacity using gadoxetic acid-enhanced HBP-MRI.

Improved Diagnostic Accuracy of Alzheimer's Disease by Combining Regional Cortical Thickness and Default Mode Network Functional Connectivity: Validated in the Alzheimer's Disease Neuroimaging Initiative Set

Park, Ji Eun,Park, Bumwoo,Kim, Sang Joon,Kim, Ho Sung,Choi, Choong Gon,Jung, Seung Chai,Oh, Joo Young,Lee, Jae-Hong,Roh, Jee Hoon,Shim, Woo Hyun The Korean Society of Radiology 2017 KOREAN JOURNAL OF RADIOLOGY Vol.18 No.6

<P><B>Objective</B></P><P>To identify potential imaging biomarkers of Alzheimer's disease by combining brain cortical thickness (CThk) and functional connectivity and to validate this model's diagnostic accuracy in a validation set.</P><P><B>Materials and Methods</B></P><P>Data from 98 subjects was retrospectively reviewed, including a study set (n = 63) and a validation set from the Alzheimer's Disease Neuroimaging Initiative (n = 35). From each subject, data for CThk and functional connectivity of the default mode network was extracted from structural T1-weighted and resting-state functional magnetic resonance imaging. Cortical regions with significant differences between patients and healthy controls in the correlation of CThk and functional connectivity were identified in the study set. The diagnostic accuracy of functional connectivity measures combined with CThk in the identified regions was evaluated against that in the medial temporal lobes using the validation set and application of a support vector machine.</P><P><B>Results</B></P><P>Group-wise differences in the correlation of CThk and default mode network functional connectivity were identified in the superior temporal (<I>p</I> < 0.001) and supramarginal gyrus (<I>p</I> = 0.007) of the left cerebral hemisphere. Default mode network functional connectivity combined with the CThk of those two regions were more accurate than that combined with the CThk of both medial temporal lobes (91.7% vs. 75%).</P><P><B>Conclusion</B></P><P>Combining functional information with CThk of the superior temporal and supramarginal gyri in the left cerebral hemisphere improves diagnostic accuracy, making it a potential imaging biomarker for Alzheimer's disease.</P>

Radiomics Analysis of Gadoxetic Acid-enhanced MRI for Staging Liver Fibrosis

Park, Hyo Jung,Lee, Seung Soo,Park, Bumwoo,Yun, Jessica,Sung, Yu Sub,Shim, Woo Hyun,Shin, Yong Moon,Kim, So Yeon,Lee, So Jung,Lee, Moon-Gyu Radiological Society of North America 2019 Radiology Vol. No.

Quantitative analysis of diffusion-weighted magnetic resonance imaging of the pancreas: Usefulness in characterizing solid pancreatic masses

Lee, Seung Soo,Byun, Jae Ho,Park, Beom Jin,Park, Seong Ho,Kim, Namkug,Park, Bumwoo,Kim, Jeong Kon,Lee, Moon-Gyu Wiley Subscription Services, Inc., A Wiley Company 2008 Journal of magnetic resonance imaging Vol.28 No.4

<B>Purpose</B><P>To evaluate whether measurement of apparent diffusion coefficient (ADC) and pure diffusion coefficient (D) can help to characterize solid pancreatic masses.</P><B>Materials and Methods</B><P>Diffusion-weighted MR imaging was performed in both a patient group (n = 71; pancreatic cancer [n = 47], mass-forming pancreatitis [n = 13], solid pseudopapillary neoplasm [n = 6], and neuroendocrine tumor [n = 5]) and a normal control group (n = 11) by applying three b-factors of 0, 500, and 1000 sec/mm<SUP>2</SUP>. ADC<SUB>500</SUB>, ADC<SUB>1000</SUB>, D (ADC using b = 500 and 1000 sec/mm<SUP>2</SUP>), and perfusion fraction (f, 1- exp [-500 sec/mm<SUP>2</SUP> × (ADC<SUB>500</SUB> – D)]) of normal pancreas, pancreatic cancer, and mass-forming pancreatitis were compared using the Kruskal–Wallis test. Receiver operating characteristic (ROC) analysis was performed to evaluate the diagnostic performance and optimal cutoff value of these parameters in differentiating pancreatic cancer from mass-forming pancreatitis.</P><B>Results</B><P>Normal pancreas had significantly higher mean ADC<SUB>500</SUB>, ADC<SUB>1000</SUB>, and f than either pancreatic cancer (P < 0.001, < 0.001, and 0.004, respectively) or mass-forming pancreatitis (P < 0.001, < 0.001, and 0.002, respectively). ADC<SUB>500</SUB>, ADC<SUB>1000</SUB>, and D of mass-forming pancreatitis were significantly lower than those of pancreatic cancer (P = 0.002, 0.004, and 0.014, respectively). Sensitivities and specificities in the diagnosis of pancreatic cancer were 72.3% and 76.9% for ADC<SUB>500</SUB>, 87.2% and 69.2% for ADC<SUB>1000</SUB>, 87.2% and 61.5% for D, and 42.6% and 92.3% for f, respectively.</P><B>Conclusion</B><P>Measurement of ADC and D may be helpful in differentiating pancreatic cancers from mass-forming pancreatitis. J. Magn. Reson. Imaging 2008;28:928–936. © 2008 Wiley-Liss, Inc.</P>

Reproducibility of measurement of apparent diffusion coefficients of malignant hepatic tumors: Effect of DWI techniques and calculation methods

Kim, So Yeon,Soo Lee, Seung,Bumwoo Park,,Kim, Namkug,Kim, Jeong Kon,Park, Seong Ho,Byun, Jae Ho,Song, Ki Jun,Koo, Ja‐,heung,Kyung Choi, Eun,Lee, Moon‐,Gyu RADIOLOGICAL SOCIETY OF NORTH AMERICA INC 2012 JOURNAL OF MAGNETIC RESONANCE IMAGING Vol.36 No.5

<P><B>Abstract</B></P><P><B>Purpose:</B></P><P>To evaluate the effect of diffusion‐weighted imaging (DWI) methods, apparent diffusion coefficient (ADC) calculation methods, and selection of b‐values on the ADCs and the measurement reproducibility of malignant hepatic tumors.</P><P><B>Materials and Methods:</B></P><P>Nineteen patients with pathologically confirmed malignant hepatic tumors underwent breath‐hold DWI (b‐values = 0, 50, 500 s/mm<SUP>2</SUP>) and respiratory‐triggered DWI (0, 50, 300, 500, 1000 s/mm<SUP>2</SUP>) twice on a 1.5 T magnetic resonance imaging (MRI) scanner. ADCs were calculated using a two b‐value and/or a multiple b‐value method. The reproducibility of the ADC measurements was evaluated from the intraclass correlation coefficients (ICCs) and the 95% Bland–Altman limit‐of‐agreement (LOA).</P><P><B>Results:</B></P><P>The ADCs were different according to the DWI methods (<I>P</I> = 0.040–0.282), ADC calculation methods (<I>P</I> = 0.003–0.825), and the choice of b‐values (<I>P</I> < 0.001). The ADC tended to be more reproducible with use of breath‐hold DWI (ICC: 0.898–0.933; LOA, 18.8%–24.0%) than respiratory‐triggered DWI (ICC: 0.684–0.928; LOA, 15.0%–31.9%) (<I>P</I> = 0.008–0.122). For respiratory‐triggered DWI, the multiple b‐value method using five b‐values had better reproducibility than the two b‐value method for measurement of ADC (<I>P</I> = 0.009–0.221).</P><P><B>Conclusion:</B></P><P>The DWI method, ADC calculation method, and selection of b‐values potentially influence the ADCs and the reproducibility of malignant hepatic tumors. ADCs calculated from breath‐hold DWI are more reproducible than from respiratory‐triggered DWI. A multiple b‐value method may improve the reproducibility of respiratory‐triggered DWI. J. Magn. Reson. Imaging 2012;36:1131–1138. © 2012 Wiley Periodicals, Inc.</P>

Prediction of Decompensation and Death in Advanced Chronic Liver Disease Using Deep Learning Analysis of Gadoxetic Acid-Enhanced MRI

Heo Subin,이승수,Kim So Yeon,Lim Young-Suk,Park Hyo Jung,Yoon Jee Seok,Suk Heung-Il,Sung Yu Sub,Park Bumwoo,Lee Ji Sung 대한영상의학회 2022 Korean Journal of Radiology Vol.23 No.12

Objective: This study aimed to evaluate the usefulness of quantitative indices obtained from deep learning analysis of gadoxetic acid-enhanced hepatobiliary phase (HBP) MRI and their longitudinal changes in predicting decompensation and death in patients with advanced chronic liver disease (ACLD). Materials and Methods: We included patients who underwent baseline and 1-year follow-up MRI from a prospective cohort that underwent gadoxetic acid-enhanced MRI for hepatocellular carcinoma surveillance between November 2011 and August 2012 at a tertiary medical center. Baseline liver condition was categorized as non-ACLD, compensated ACLD, and decompensated ACLD. The liver-to-spleen signal intensity ratio (LS-SIR) and liver-to-spleen volume ratio (LS-VR) were automatically measured on the HBP images using a deep learning algorithm, and their percentage changes at the 1-year follow-up (ΔLS-SIR and ΔLS-VR) were calculated. The associations of the MRI indices with hepatic decompensation and a composite endpoint of liver-related death or transplantation were evaluated using a competing risk analysis with multivariable Fine and Gray regression models, including baseline parameters alone and both baseline and follow-up parameters. Results: Our study included 280 patients (153 male; mean age ± standard deviation, 57 ± 7.95 years) with non-ACLD, compensated ACLD, and decompensated ACLD in 32, 186, and 62 patients, respectively. Patients were followed for 11–117 months (median, 104 months). In patients with compensated ACLD, baseline LS-SIR (sub-distribution hazard ratio [sHR], 0.81; p = 0.034) and LS-VR (sHR, 0.71; p = 0.01) were independently associated with hepatic decompensation. The ∆LS-VR (sHR, 0.54; p = 0.002) was predictive of hepatic decompensation after adjusting for baseline variables. ∆LS-VR was an independent predictor of liver-related death or transplantation in patients with compensated ACLD (sHR, 0.46; p = 0.026) and decompensated ACLD (sHR, 0.61; p = 0.023). Conclusion: MRI indices automatically derived from the deep learning analysis of gadoxetic acid-enhanced HBP MRI can be used as prognostic markers in patients with ACLD.

Netrin‐1/ DCC ‐mediated PLC γ1 activation is required for axon guidance and brain structure development

Kang, Du‐,Seock,Yang, Yong Ryoul,Lee, Cheol,Park, BumWoo,Park, Kwang IL,Seo, Jeong Kon,Seo, Young Kyo,Cho, HyungJoon,Cocco, Lucio,Suh, Pann‐,Ghill EMBO 2018 EMBO reports Vol.19 No.11