• Journal of Korea Multimedia Society
• /
• v.24 no.3
• /
• pp.373-381
• /
• 2021

In this paper, we propose a classification model based on convolutional neural network(CNN) for predicting 2-year recurrence in non-small cell lung cancer(NSCLC) patients using preoperative chest CT images. Based on the region of interest(ROI) defined as the tumor internal and external area, the input images consist of an intratumoral patch, a peritumoral patch and a peritumoral texture patch focusing on the texture information of the peritumoral patch. Each patch is trained through AlexNet pretrained on ImageNet to explore the usefulness and performance of various patches. Additionally, ensemble learning of network trained with each patch analyzes the performance of different patch combination. Compared with all results, the ensemble model with intratumoral and peritumoral patches achieved the best performance (ACC=98.28%, Sensitivity=100%, NPV=100%).

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.14
• /
• pp.5539-5544
• /
• 2014

Background: To evaluate the location of tumor relapse and imaging modality for detection according to the breast cancer subtype: luminal A, luminal B, HER2 positive luminal B, nonluminal HER2 positive, and triple negative. Materials and Methods: A total of 1244 patients with breast cancer with known estrogen receptor (ER), progesterone receptor (PR), Ki-67 and human epidermal growth factor receptor 2 (HER2), who underwent breast surgery from 2009 to 2012 were analyzed. Patients were classified into the following categories: luminal A (n=458), luminal B (n=241), HER2 positive luminal B (n=227), nonluminal HER2 positive (n=145) and triple negative (n=173). A total of 105 cases of relapse were detected in 102 patients: locoregional recurrence (n=46), recurrence in the contralateral breast (n=28) and distant metastasis (n=31). Comparison of proportions was used to determine the difference between subtypes. Results: Relapse rates by subtypes are as follows: luminal A 23 of 458 (5.02%), luminal B 19 of 241(7.88%), HER2 positive luminal B 15 of 227 (6.61%), nonluminal HER2 postive 19 of 145 (13.10%) and triple negative 29 of 173(16.76%). Luminal A tumors had the lowest rate of recurrence and had significantly lower recurrence rate in comparison with nonluminal HER2 postive (p=0.0017) and triple negative subtypes (p<0.0001). Compared with all other subtypes except nonluminal HER2 positive, triple negative tumors had the highest rate of tumor recurrence (p<0.01). Triple negatives were most likely to develop contralateral recurrence against all subtypes (p<0.05). Detection rate of locoregional and contralateral tumor recurrence were 28.3% on mammography (n=17/60). Conclusions: Luminal A tumors are associated with a low risk of recurrence while triple negative lesions have a high risk. In case of triple negative tumors, the contralateral breast has much more recurrence as compared with all other subtype. In terms of detection rates, breast USG was the best modality for detecting tumor recurrence, compared with other modalities (p<0.05). Subtyping of breast tumors using a molecular gene expression panel can identify patients who have increased risk of recurrence and allow prediction of locations of tumor recurrence for each subtype.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.23
• /
• pp.10401-10405
• /
• 2015

Background: Molecular prognostic markers have been under investigation for the last decade and no validated marker to date has been proven to be used in daily clinical practice for urinary bladder cancers. The aim of the present study is to evaluate the significance of HYAL-1 expression in prediction of recurrence and progression in pT1 urothelial carcinomas. Materials and Methods: Eighty-nine urothelial carcinoma cases staged as T1 according to 2004 WHO classification were studied. Representative sections from every case were stained immunohistochemically for HYAL-1 and scored between 0 and +3, according to staining density, and graded as low and high for the scores 0-1 and 2-3, respectively. Results: Of the 89 pT1 bladder cancer patients, HYAL-1 expression was high in 92.1% (82 patients; 72 patients +3 and 10 patients +2) and low in 7.9% (only 7 patients; 6 patients +1 and 1 patient 0) of the cases. Of the 89 patients, 38 (42.7%) had recurrence and 22 (24.7%) showed progression. HYAL-1 staining did not show significant characteristics for tumor grade, accompanying CIS, multiplicity, tumor size, age and sex. HYAL-1 expression did not have any prognostic value in estimating recurrence or progression. Conclusions: HYAL-1 expression was found to be high, but did not have any prognostic importance in T1 bladder urothelial carcinomas.

• Journal of Digestive Cancer Research
• /
• v.3 no.2
• /
• pp.89-94
• /
• 2015

The role of adjuvant chemotherapy in patients with stage II colon cancer remains a controversial issue. Adjuvant chemotherapy aims to eliminate any micrometastatic disease that may have been missed, at the time of surgery. Although one prospective study showed a small but statistically significant benefit with respect to the overall survival for those who received adjuvant chemotherapy, multiple pooled data did not demonstrate any benefit of this therapy in patients with stage II colon cancer. Current national and international guidelines for the adjuvant treatment of stage II colon dose not advise routine implementation of adjuvant chemotherapy, but rather recommend selective use of this therapy for patients with high risk of recurrence. High risk features for recurrence include T4 disease, poorly differentiated histology, presence of lymphovascular invasion, presence of perineural invasion, inadequate retrieval of lymph nodes, bowel obstruction, localized perforation, or positive margins. More recently, prediction tools using gene expression cancer profiles are proposed to identify patients who are most likely to have recurrence and therefore may benefit from postoperative chemotherapy in stage II colon cancer. These novel methods together with conventional prognosticators, will allow us to implement more optimized personalizing adjuvant therapy in these patients.

• Journal of the Korean Statistical Society
• /
• v.3 no.2
• /
• pp.103-115
• /
• 1974

During the last few years there has been an interest in models for qualitative attributes, where complex signifies that affection may be caused in two or more ways [1-3]. These models have in common the prediction of variable recurrence risks among families with given parental phenotpes. Segregation analysis has covered only a few cases [4,5]. The present paper extends segregation analysis to three complex models under two mode of ascertainment.

• Journal of Gastric Cancer
• /
• v.5 no.4 s.20
• /
• pp.266-272
• /
• 2005

Purpose: Recurrence occurs in various forms and in different organs after a curative resection of gastric cancer. The most common location for hematogenous recurrence is the liver. The aim of this study was to evaluate the clinicopathological risk factors associated with hepatic recurrence after surgical treatment of gastric cancer. Materials and Methods: We reviewed the medical records of 838 patients who had taken radical surgery for primary gastric cancers at Chonbuk National University Hospital between January 1992 and December 1999. According to the medical records, we retrospectively investigated the association between the clinicopathologic variables and hepatic recurence. Results: Recurrence of gastric cancer was documented in 201 out of the 838 patients (23.98%). Hepatic recurrences were found in 59 out of 201 patients (29.35%). Multiple logistic regression analysis identified the following as independent risk factors of hepatic recurrence: Lauren's Intestinal type (OR, 6.60; 95% Cl, 1.53 to 28.9; p=0.011) and proximal resection of margin below 6 cm (OR, 3.76; 95% Cl, 1.03 to 13.67; P=0.045). Conclusion: Various studies on clinocopathologic risk factors of liver recurrence with molecular biologic research should make possible the prediction of recurrence and help high-risk patients to find appropriate management.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.14 no.2
• /
• pp.276-287
• /
• 1990

In this paper research result of transient thermal stress analysis of power plant turbine rotors for life prediction under severs operating conditions is presented. Galerkin's recurrence scheme is used for numerical solution of discretized FEM equation of transient heat conduction equation. Boundary conditions for the equation and operating conditions are intensively investigated for accurate life prediction of turbine rotors in operation. A computer program for on-site application is developed and tested. Distribution of thermal stress in turbine rotors during various operating condition is analyzed with the program and it is found that the peak thermal stress appears during cold stage conditions at the first stage of high pressure rotors.

• Korean Journal of Radiology
• /
• v.24 no.4
• /
• pp.338-348
• /
• 2023

Objective: Patients with a history of ischemic stroke are at risk for a second ischemic stroke. This study aimed to investigate the relationship between carotid plaque enhancement on perfluorobutane microbubble contrast-enhanced ultrasonography (CEUS) and future recurrent stroke, and to determine whether plaque enhancement can contribute to risk assessment for recurrent stroke compared with the Essen Stroke Risk Score (ESRS). Materials and Methods: This prospective study screened 151 patients with recent ischemic stroke and carotid atherosclerotic plaques at our hospital between August 2020 and December 2020. A total of 149 eligible patients underwent carotid CEUS, and 130 patients who were followed up for 15-27 months or until stroke recurrence were analyzed. Plaque enhancement on CEUS was investigated as a possible risk factor for stroke recurrence and as a possible adjunct to ESRS. Results: During follow-up, 25 patients (19.2%) experienced recurrent stroke. Patients with plaque enhancement on CEUS had an increased risk of stroke recurrence events (22/73, 30.1%) compared to those without plaque enhancement (3/57, 5.3%), with an adjusted hazard ratio (HR) of 38.264 (95% confidence interval [CI]:14.975-97.767; P < 0.001) according to a multivariable Cox proportional hazards model analysis, indicating that the presence of carotid plaque enhancement was a significant independent predictor of recurrent stroke. When plaque enhancement was added to the ESRS, the HR for stroke recurrence in the high-risk group compared to that in the low-risk group (2.188; 95% CI, 0.025-3.388) was greater than that of the ESRS alone (1.706; 95% CI, 0.810-9.014). A net of 32.0% of the recurrence group was reclassified upward appropriately by the addition of plaque enhancement to the ESRS. Conclusion: Carotid plaque enhancement was a significant and independent predictor of stroke recurrence in patients with ischemic stroke. Furthermore, the addition of plaque enhancement improved the risk stratification capability of the ESRS.

• Nuclear Medicine and Molecular Imaging
• /
• v.42 no.sup1
• /
• pp.52-59
• /
• 2008

In the management of colo-retal and anal cancer, accurate staging, treatment evaluation, early detection of recurrence are main clinical problems. F-18 FDG PET (PET/CT) has been reported as useful in the management of colo-rectal and anal cancer because that PET has high diagnostic performance comparing to conventional studies. In case of liver metastases, for confirmation of no extrahepatic metastases, in case of high risk of metastasis, for avoiding unnecessary operation, PET (PET/CT) is expected more useful. In anal cancer, PET is expected useful in lymph node staging. For the early prediction of chemotherapy or radiation therapy effect PET has been reported as useful, also. In early detection of recurrence by PET, cost-benefit advantages has been suggested, also. PET/CT is expected to have higher diagnostic performance than PET alone.

• Journal of the Korean Society of Radiology
• /
• v.81 no.5
• /
• pp.1164-1174
• /
• 2020

Purpose To evaluate a deep learning model to predict recurrence of thyroid tumor using preoperative ultrasonography (US). Materials and Methods We included representative images from 229 US-based patients (male:female = 42:187; mean age, 49.6 years) who had been diagnosed with thyroid cancer on preoperative US and subsequently underwent thyroid surgery. After selecting each representative transverse or longitudinal US image, we created a data set from the resulting database of 898 images after augmentation. The Python 2.7.6 and Keras 2.1.5 framework for neural networks were used for deep learning with a convolutional neural network. We compared the clinical and histological features between patients with and without recurrence. The predictive performance of the deep learning model between groups was evaluated using receiver operating characteristic (ROC) analysis, and the area under the ROC curve served as a summary of the prognostic performance of the deep learning model to predict recurrent thyroid cancer. Results Tumor recurrence was noted in 49 (21.4%) among the 229 patients. Tumor size and multifocality varied significantly between the groups with and without recurrence (p < 0.05). The overall mean area under the curve (AUC) value of the deep learning model for prediction of recurrent thyroid cancer was 0.9 ± 0.06. The mean AUC value was 0.87 ± 0.03 in macrocarcinoma and 0.79 ± 0.16 in microcarcinoma. Conclusion A deep learning model for analysis of US images of thyroid cancer showed the possibility of predicting recurrence of thyroid cancer.