• Korean Journal of Radiology
• /
• v.23 no.12
• /
• pp.1269-1280
• /
• 2022

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 ΔLS-VR (sHR, 0.54; p = 0.002) was predictive of hepatic decompensation after adjusting for baseline variables. ΔLS-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.

• Korean Journal of Radiology
• /
• v.23 no.12
• /
• pp.1281-1289
• /
• 2022

Objective: Radiomic modeling using multiple regions of interest in MRI of the brain to diagnose juvenile myoclonic epilepsy (JME) has not yet been investigated. This study aimed to develop and validate radiomics prediction models to distinguish patients with JME from healthy controls (HCs), and to evaluate the feasibility of a radiomics approach using MRI for diagnosing JME. Materials and Methods: A total of 97 JME patients (25.6 ± 8.5 years; female, 45.5%) and 32 HCs (28.9 ± 11.4 years; female, 50.0%) were randomly split (7:3 ratio) into a training (n = 90) and a test set (n = 39) group. Radiomic features were extracted from 22 regions of interest in the brain using the T1-weighted MRI based on clinical evidence. Predictive models were trained using seven modeling methods, including a light gradient boosting machine, support vector classifier, random forest, logistic regression, extreme gradient boosting, gradient boosting machine, and decision tree, with radiomics features in the training set. The performance of the models was validated and compared to the test set. The model with the highest area under the receiver operating curve (AUROC) was chosen, and important features in the model were identified. Results: The seven tested radiomics models, including light gradient boosting machine, support vector classifier, random forest, logistic regression, extreme gradient boosting, gradient boosting machine, and decision tree, showed AUROC values of 0.817, 0.807, 0.783, 0.779, 0.767, 0.762, and 0.672, respectively. The light gradient boosting machine with the highest AUROC, albeit without statistically significant differences from the other models in pairwise comparisons, had accuracy, precision, recall, and F1 scores of 0.795, 0.818, 0.931, and 0.871, respectively. Radiomic features, including the putamen and ventral diencephalon, were ranked as the most important for suggesting JME. Conclusion: Radiomic models using MRI were able to differentiate JME from HCs.

• Korean Journal of Radiology
• /
• v.23 no.3
• /
• pp.333-342
• /
• 2022

Objective: Intracranial atherosclerotic stroke occurs through various mechanisms, mainly by artery-to-artery embolism (AA) or branch occlusive disease (BOD). This study evaluated the spatial relationship between middle cerebral artery (MCA) plaques and perforating arteries among different MCA territory infarction types using vessel wall magnetic resonance imaging (VW-MRI). Materials and Methods: We retrospectively enrolled patients with acute MCA infarction who underwent VW-MRI. Thirty-four patients were divided into three groups according to infarction pattern: 1) BOD, 2) both BOD and AA (BOD-AA), and 3) AA. To determine the factors related to BOD, the BOD and BOD-AA groups were combined into one group (with striatocapsular infarction [BOD+]) and compared with the AA group. To determine the factors related to AA, the BOD-AA and AA groups were combined into another group (with cortical infarction [AA+]) and compared with the BOD group. Plaque morphology and the spatial relationship between the perforating artery orifice and plaque were evaluated both quantitatively and qualitatively. Results: The plaque margin in the BOD+ group was closer to the perforating artery orifice than that in the AA group (p = 0.011), with less enhancing plaque (p = 0.030). In the BOD group, plaques were mainly located on the dorsal (41.2%) and superior (41.2%) sides where the perforating arteries mainly arose. No patient in the AA group had overlapping plaques with perforating arteries at the cross-section where the perforator arose. Perforating arteries associated with culprit plaques were most frequently located in the middle two-thirds of the M1 segment (41.4%). The AA+ group had more stenosis (%) than the BOD group (39.73 ± 24.52 vs. 14.42 ± 20.96; p = 0.003). Conclusion: The spatial relationship between the perforating artery orifice and plaque varied among different types of MCA territory infarctions. In patients with BOD, the plaque margin was closer and blocked the perforating artery orifice, and stenosis degree and enhancement were less than those in patients with AA.

• Journal of Korean Neurosurgical Society
• /
• v.67 no.2
• /
• pp.237-248
• /
• 2024

Objective : Postoperative data on Cushing's disease (CD) are equivocal in the literature. These discrepancies may be attributed to different series with different criteria for remission and variable follow-up durations. Additional data from experienced centers may address these discrepancies. In this study, we present the results obtained from 96 endoscopic transsphenoidal surgeries (ETSSs) for CD conducted in a well-experienced center. Methods : Pre- and postoperative data of 96 ETSS in 87 patients with CD were included. All cases were handled by the same neurosurgical team between 2014 and 2022. We obtained data on remission status 3-6 months postoperatively (medium-term) and during the latest follow-up (long-term). Additionally, magnetic resonance imaging (MRI) and pathology results were obtained for each case. Results : The mean follow-up duration was 39.5±3.2 months. Medium and long-term remission rates were 77% and 82%, respectively. When only first-time operations were considered, the medium- and long-term remission rates were 78% and 82%, respectively. The recurrence rate in this series was 2.5%. Patients who showed remission between 3-6 months had higher long-term remission rates than did those without initial remission. Tumors >2 cm and extended tumor invasion of the cavernous sinus (Knosp 4) were associated with lower postoperative remission rates. Conclusion : Adenoma size and the presence/absence of cavernous sinus invasion on preopera-tive MRI may predict long-term postoperative remission. A tumor size of 2 cm may be a supporting criterion for predicting remission in Knosp 4 tumors. Further studies with larger patient populations are necessary to support this finding.

• Journal of the Korean Society of Radiology
• /
• v.18 no.1
• /
• pp.37-44
• /
• 2024

Even though MR can reveal excellent soft-tissue contrast and functional information, CT is also required for electron density information for accurate dose calculation in Radiotherapy. For the fusion of MRI and CT images in RT treatment planning workflow, patients are normally scanned on both MRI and CT imaging modalities. Recently deep-learning-based generations of CT images from MR images became possible owing to machine learning technology. This eliminated CT scanning work. This study implemented a CycleGan deep-learning-based CT image generation from MR images. Three CT generators whose learning is based on T1- , T2- , or T1-&T2-weighted MR images were created, respectively. We found that the T1-weighted MR image-based generator can generate better than other CT generators when T1-weighted MR images are input. In contrast, a T2-weighted MR image-based generator can generate better than other CT generators do when T2-weighted MR images are input. The results say that the CT generator from MR images is just outside the practical clinics and the specific weight MR image-based machine-learning generator can generate better CT images than other sequence MR image-based generators do.

• Journal of the Korean Society of Radiology
• /
• v.84 no.4
• /
• pp.970-976
• /
• 2023

This study reports on diffuse leptomeningeal glioneuronal tumor (DL-GNT) in a 29- year-old male. DL-GNT is a rare central nervous system (CNS) tumor mostly seen in children and only few cases have been reported in adult patients. Our patient presented with a chronic headache that lasted for five months. MR imaging showed mild hydrocephalus, multiple rim-enhancing nodular lesions in the suprasellar cistern, diffuse leptomeningeal enhancement in the lumbosacral area, and multiple small non-enhancing cyst-appearing lesions not suppressed on fluid attenuated inversion recovery (FLAIR) images in the bilateral basal ganglia, thalami, and cerebral hemispheres. Under the impression of germ cell tumor with leptomeningeal seeding, the patient underwent trans-sphenoidal tumor removal. DL-GNT was pathologically confirmed and FGFR1 mutation was detected through a next-generation sequencing test. In conclusion, a combination of leptomeningeal enhancement and multiple parenchymal non-enhancing cyst-appearing lesions not suppressed on FLAIR images may be helpful for differential diagnosis despite overlapping imaging features with many other CNS diseases that have leptomeningeal enhancement.

• Journal of the Korean Society of Radiology
• /
• v.84 no.6
• /
• pp.1361-1366
• /
• 2023

The persistent primitive olfactory artery (PPOA) is a rare variant of the anterior cerebral artery, first reported in 1979. It reportedly has a high correlation with the development of aneurysms, owing to the hemodynamic stress induced by the structural characteristics of the hairpin turn. Herein, we present a rare case of PPOA type 4 with a fusiform aneurysm at the hairpin turn segment in a 46-year-old female with occasional headaches. Time-of-flight MR angiography and transfemoral cerebral angiography revealed an unusual branch arising from the left A1 segment, running anteromedially along the ipsilateral olfactory tract, and turning the hairpin posterior to the olfactory bulb. This branch continued into the left accessory middle cerebral artery, and a fusiform aneurysm was observed at the hairpin segment. No further treatment was performed, and follow-up imaging was recommended. Nevertheless, it is essential to recognize and diagnose these rare variations.

• Journal of the Korean Society of Radiology
• /
• v.84 no.5
• /
• pp.1080-1090
• /
• 2023

Purpose This study aimed to compare the volume and normative percentiles of brain volumetry in the Korean population using quantitative brain volumetric MRI analysis tools NeuroQuant^Ⓡ (NQ) and DeepBrain^Ⓡ (DB), and to evaluate whether the differences in the normative percentiles of brain volumetry between the two tools is related to cranial shape. Materials and Methods In this retrospective study, we analyzed the brain volume reports obtained from NQ and DB in 163 participants without gross structural brain abnormalities. We measured threedimensional diameters to evaluate the cranial shape on T1-weighted images. Statistical analyses were performed using intra-class correlation coefficients and linear correlations. Results The mean normative percentiles of the thalamus (90.8 vs. 63.3 percentile), putamen (90.0 vs. 60.0 percentile), and parietal lobe (80.1 vs. 74.1 percentile) were larger in the NQ group than in the DB group, whereas that of the occipital lobe (18.4 vs. 68.5 percentile) was smaller in the NQ group than in the DB group. We found a significant correlation between the mean normative percentiles obtained from the NQ and cranial shape: the mean normative percentile of the occipital lobe increased with the anteroposterior diameter and decreased with the craniocaudal diameter. Conclusion The mean normative percentiles obtained from NQ and DB differed significantly for many brain regions, and these differences may be related to cranial shape.

• Journal of the Korean Society of Radiology
• /
• v.84 no.6
• /
• pp.1290-1308
• /
• 2023

Purpose To compare the diagnostic performance of rectal CT with that of high-resolution rectal MRI and histopathology in assessing rectal cancer. Materials and Methods Sixty-seven patients with rectal cancer who underwent rectal CT with rectal distension using sonographic gel and high-resolution MRI were enrolled in this study. The distance from the anal verge/anorectal junction, distance to the mesorectal fascia (MRF), extramural depth (EMD), extramesorectal lymph node (LN) involvement, extramural venous invasion (EMVI), and T/N stages in rectal CT/MRI were analyzed by two gastrointestinal radiologists. The CT findings of 20 patients who underwent radical surgery without concurrent chemoradiotherapy were compared using histopathology. Interclass correlations and kappa statistics were used. Results The distance from the anal verge/anorectal junction showed an excellent intraclass correlation between CT and MRI for both reviewers. For EMD, the distance to the MRF, presence of LNs, extramesorectal LN metastasis, EMVI, T stage, and intermodality kappa or weighted kappa values between CT and MRI showed excellent agreement. Among the 20 patients who underwent radical surgery, T staging, circumferential resection margin involvement, EMVI, and LN metastasis on rectal CT showed acceptable concordance rates with histopathology. Conclusion Dedicated rectal CT may be on par with rectal MRI in providing critical information to patients with rectal cancer.

• Journal of the Korean Society of Radiology
• /
• v.85 no.3
• /
• pp.654-660
• /
• 2024

Primary malignant fibrous histiocytoma (MFH) is a malignant tumor of mesenchymal origin that rarely occurs in the urinary tract, particularly in the urinary bladder. Unlike urothelial carcinoma, which accounts for most bladder cancers, it occurs in the submucosal portion of the bladder wall and consists of the lamina propria, muscularis propria, and adventitia. It is presumed to originate from poorly differentiated pluripotent mesenchymal cells in which fibroblasts and histiocytes are partially differentiated. Radiologically, it is known as the "non-papillary tumor" and is commonly diagnosed as a large mass without necrosis, which shows invasion beyond the muscularis propia. Although the prognosis of this rare malignancy depends on pathological parameters, it generally has a poor prognosis with high local tumor recurrence. Here, we present a case of primary MFH in the urinary bladder with clinical symptoms of lower abdominal pain without gross hematuria that recurred rapidly and showed an aggressive disease course.