• Journal of Digital Convergence
• /
• v.12 no.3
• /
• pp.367-375
• /
• 2014

In this thesis we observe microvascular structure in mice by using micro-computed tomography (CT), which is high-resolution X-ray imaging equipment that can acquire Real-time dynamic image, and it aims to investigate the usefulness of micro-CT developed by Institute for Radiological Imaging Science Wonkwang University School of Medicine. After acquiring the systemic images of rats, contrast-enhanced 3D images of vascular structures could be acquired by using Maximum Intensity Projection (MIP) and Volume Rending Technique (VRT), This was divided into each vascular system of head, abdomen and heart and systemic vascular system.

• Korean Journal of Radiology
• /
• v.22 no.3
• /
• pp.344-353
• /
• 2021

Objective: The mitotic count of gastrointestinal stromal tumors (GIST) is closely associated with the risk of planting and metastasis. The purpose of this study was to develop a predictive model for the mitotic index of local primary GIST, based on deep learning algorithm. Materials and Methods: Abdominal contrast-enhanced CT images of 148 pathologically confirmed GIST cases were retrospectively collected for the development of a deep learning classification algorithm. The areas of GIST masses on the CT images were retrospectively labelled by an experienced radiologist. The postoperative pathological mitotic count was considered as the gold standard (high mitotic count, > 5/50 high-power fields [HPFs]; low mitotic count, ≤ 5/50 HPFs). A binary classification model was trained on the basis of the VGG16 convolutional neural network, using the CT images with the training set (n = 108), validation set (n = 20), and the test set (n = 20). The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated at both, the image level and the patient level. The receiver operating characteristic curves were generated on the basis of the model prediction results and the area under curves (AUCs) were calculated. The risk categories of the tumors were predicted according to the Armed Forces Institute of Pathology criteria. Results: At the image level, the classification prediction results of the mitotic counts in the test cohort were as follows: sensitivity 85.7% (95% confidence interval [CI]: 0.834-0.877), specificity 67.5% (95% CI: 0.636-0.712), PPV 82.1% (95% CI: 0.797-0.843), NPV 73.0% (95% CI: 0.691-0.766), and AUC 0.771 (95% CI: 0.750-0.791). At the patient level, the classification prediction results in the test cohort were as follows: sensitivity 90.0% (95% CI: 0.541-0.995), specificity 70.0% (95% CI: 0.354-0.919), PPV 75.0% (95% CI: 0.428-0.933), NPV 87.5% (95% CI: 0.467-0.993), and AUC 0.800 (95% CI: 0.563-0.943). Conclusion: We developed and preliminarily verified the GIST mitotic count binary prediction model, based on the VGG convolutional neural network. The model displayed a good predictive performance.

• Tuberculosis and Respiratory Diseases
• /
• v.44 no.2
• /
• pp.264-279
• /
• 1997

Background : To analyze the morphologic characteristics of low density lymph node in etiologic differentiation of lymphadenopathy, emphasizing the different features between tuberculosis and lung cancer, on contrast enhanced CT scan. Method : A total of 64 patients who showed low density lymph nodes on chest CT scan were analyzed. Primary causes were tuberculosis (n=28), lung cancer (n=27), malignant lymphoma (n=5) and metastasis from extrathoracic malignancies (n=4). CT scan was performed with 10mm slice thickness and 7 characteristic features were evaluated : location, size, presence or absence of the nonnecrotic lymph node, calcification, perinodal fat obliteration, thickness and evenness of the enhancing rim. Results : In patients with tuberculous lymphadenopathy, lymph nodes with uneven (68.0%) and thick (62.1%) enhancing rim were more common than lung cancer (p<0.05). Low density lymph nodes with less than 1 cm in size were found only in tuberculous lymphadenopathy(n=10). In 48.2% of patients with lung cancer, more than 1 nonnecrotic enlarged lymph node were coexisted, whereas 21.4% in patients with tuberculous lymphadenopathy(p=0.06). However, the size, location and calcification were not statistically significant between tuberculous lymphadenopathy and lung cancer. Conclusion : Tuberculous lymphadenopathy is strongly suggested when enhancing rim of enlarged lymph nodes is uneven and thick, when the coexisting nonnecrotic lymph nodes are few in number and when central low density is encountered in normal sized lymph nodes.

• Radiation Oncology Journal
• /
• v.24 no.4
• /
• pp.248-254
• /
• 2006

$\underline{Purpose}$: To evaluate acute toxicities in cervix cancer patients receiving intensity modulated whole pelvic radiation therapy (IM-WPRT). $\underline{Materials\;and\;Methods}$: Between August 2004 and April 2006, 17 patients who underwent IM-WPRT were analysed. An intravenous contrast agent was used for radiotherapy planning computed tomography (CT). The central clinical target volume (CTV) included the primary tumor, uterus, vagina, and parametrium. The nodal CTV was defined as the lymph nodes larger than 1 cm seen on CT and the contrased-enhanced pelvic vessels. The planning target volume (PTV) was the 1-cm expanded volume around the central CTV, except for a 5-mm expansion from the posterior vagina, and the nodal PTV was defined as the nodal CTV plus a 1.5 cm margin. IM-WPRT was prescribed to deliver a dose of 50 Gy to more than 95% of the PTV. Acute toxicity was assessed with common toxicity criteria up to 60 days after radiotherapy. $\underline{Results}$: Grade 1 nausea developed in 10 (58.9%) patients, and grade 1 and 2 diarrhea developed in 11 (64.7%) and 1 (5.9%) patients, respectively. No grade 3 or higher gastrointestinal toxicity was seen. Leukopenia, anemia, and thrombocytopenia occurred in 15 (88.2%). 7 (41.2%), and 2 (11.8%) patients, respectively, as hematologic toxicities. Grade 3 leukopenia developed in 2 patients who were treated with concurrent chemoradiotherapy. $\underline{Conclusion}$: IM-WPRT can be a useful treatment for cervix cancer patients with decreased severe acute toxicities and a resultant improved compliance to whole pelvic irradiation.

• Journal of the Korean Society of Radiology
• /
• v.81 no.6
• /
• pp.1436-1447
• /
• 2020

Purpose To investigate the clinical and CT features at admission to predict the progression to necrotizing pancreatitis (NP) in patients initially diagnosed with interstitial edematous pancreatitis (IEP). Materials and Methods Patients with IEP who underwent contrast-enhanced CT at admission and follow-up CT (< 14 days) were included (n = 178). Two radiologists performed a consensus review of follow-up CT scans and diagnosed the type of acute pancreatitis as IEP or NP. Laboratory findings at admission were recorded. Clinical, CT, and laboratory findings were compared between the IEP-IEP group and IEP-NP group using the chi-square test and the t-test. Multivariate analysis was also performed. Results There were 112 and 66 patients in the IEP-IEP and the IEP-NP groups, respectively. The proportion of patients with alcohol etiology was significantly larger in the IEP-NP group. Among the CT findings, the presence of peripancreatic fluid and heterogeneous parenchymal enhancement were more frequently observed in the IEP-NP group. Among the laboratory variables, serum C-reactive protein levels and white blood cell counts were significantly higher in the IEP-NP group. Multivariate analysis revealed that the presence of peripancreatic fluid and heterogeneous parenchymal enhancement were significant findings distinguishing the two groups. Conclusion CT findings, such as the presence of peripancreatic fluid and heterogeneous pancreatic parenchymal enhancement, may be helpful in predicting the progression to NP in patients initially diagnosed with IEP.