• Korean Journal of Radiology
• /
• v.25 no.6
• /
• pp.550-558
• /
• 2024

Hepatocellular carcinoma (HCC) is a biologically heterogeneous tumor characterized by varying degrees of aggressiveness. The current treatment strategy for HCC is predominantly determined by the overall tumor burden, and does not address the diverse prognoses of patients with HCC owing to its heterogeneity. Therefore, the prognostication of HCC using imaging data is crucial for optimizing patient management. Although some radiologic features have been demonstrated to be indicative of the biologic behavior of HCC, traditional radiologic methods for HCC prognostication are based on visually-assessed prognostic findings, and are limited by subjectivity and inter-observer variability. Consequently, artificial intelligence has emerged as a promising method for image-based prognostication of HCC. Unlike traditional radiologic image analysis, artificial intelligence based on radiomics or deep learning utilizes numerous image-derived quantitative features, potentially offering an objective, detailed, and comprehensive analysis of the tumor phenotypes. Artificial intelligence, particularly radiomics has displayed potential in a variety of applications, including the prediction of microvascular invasion, recurrence risk after locoregional treatment, and response to systemic therapy. This review highlights the potential value of artificial intelligence in the prognostication of HCC as well as its limitations and future prospects.

• Fire Science and Engineering
• /
• v.31 no.1
• /
• pp.26-35
• /
• 2017

Since the Northridge earthquake (1994) and Kobe earthquake (1995), the concept of performance-based design has been actively introduced to design major structures and buildings. Recently, the seismic design code was established for fire protection facilities. Therefore, the important fire protection facilities should be designed and constructed according to the seismic design code. Accordingly, uniform hazard spectra (UHS), with annual exceedance probabilities, corresponding to the performance level, such as operational, immediate occupancy, life safety, and collapse prevention, are required for performance-based design. Using the method of probabilistic seismic hazard analysis (PSHA), the uniform hazard spectra for 5 major cities in Korea with a recurrence period of 500, 1,000, and 2,500 years corresponding to frequencies of (0.5, 1.0, 2.0, 5.0, 10.0)Hz and PGA, were analyzed. The expert panel was comprised of 10 members in seismology and tectonics. The ground motion prediction equations and several seismo tectonic models suggested by 10 expert panel members in seismology and tectonics were used as the input data for uniform hazard spectrum analysis. According to sensitivity analysis, the parameter of spectral ground motion prediction equations has a greater impact on the seismic hazard than seismotectonic models. The resulting uniform hazard spectra showed maximum values of the seismic hazard at a frequency of 10Hz and also showed the shape characteristics, which are similar to previous studies and related technical guides for nuclear facilities.

• Journal of the Korean earth science society
• /
• v.37 no.3
• /
• pp.162-172
• /
• 2016

Since the Northridge earthquake in 1994 and the Kobe earthquake in 1995 occurred, the concept of performance based design has been introduced for designing various kinds of important structures and buildings. Uniform hazard spectra (UHS), with annual exceedance probabilities, corresponding to the performance level of each structure, are required for performance-based design. The probabilistic seismic hazard analysis was performed using spectral ground motion prediction equations, which were developed from both Korean Peninsula and Central and Eastern US region, and several seismotectonic models suggested by 10 expert panel members in seismology and tectonics. The uniform hazard spectra for 5 highly populated cities in Korea, with recurrence period of 500, 1,000, and 2,500 years using the seismic hazard at the frequencies of 0.5, 1.0, 2.0, 5.0, 10.0 Hz and Peak ground acceleration (PGA) were analyzed using the probabilistic seismic hazard analysis. The sensitivity analysis suggests that spectral ground motion prediction equations impact much more on seismic hazard than what seismotectonic models do. The uniform hazard spectra commonly showed a maximum hazard at the frequency of 10 Hz and also showed the similar shape characteristics to the previous study and related technical guides to nuclear facilities.

• Journal of Gastric Cancer
• /
• v.22 no.2
• /
• pp.120-134
• /
• 2022

Purpose: This study aimed to identify prognostic factors for patients with distant lymph node-involved gastric cancer (GC) using a machine learning algorithm, a method that offers considerable advantages and new prospects for high-dimensional biomedical data exploration. Materials and Methods: This study employed 79 features of clinical pathology, laboratory tests, and therapeutic details from 289 GC patients whose distant lymphadenopathy was presented as the first episode of recurrence or metastasis. Outcomes were measured as any-cause death events and survival months after distant lymph node metastasis. A prediction model was built based on possible outcome predictors using a random survival forest algorithm and confirmed by 5×5 nested cross-validation. The effects of single variables were interpreted using partial dependence plots. A contour plot was used to visually represent survival prediction based on 2 predictive features. Results: The median survival time of patients with GC with distant nodal metastasis was 9.2 months. The optimal model incorporated the prealbumin level and the prothrombin time (PT), and yielded a prediction error of 0.353. The inclusion of other variables resulted in poorer model performance. Patients with higher serum prealbumin levels or shorter PTs had a significantly better prognosis. The predicted one-year survival rate was stratified and illustrated as a contour plot based on the combined effect the prealbumin level and the PT. Conclusions: Machine learning is useful for identifying the important determinants of cancer survival using high-dimensional datasets. The prealbumin level and the PT on distant lymph node metastasis are the 2 most crucial factors in predicting the subsequent survival time of advanced GC.

• Radiation Oncology Journal
• /
• v.26 no.3
• /
• pp.135-141
• /
• 2008

Purpose: To evaluate the clinical outcomes and prognostic factors in retroperitoneal soft tissue sarcomas treated by postoperative radiotherapy. Materials and Methods: The records of 23 patients with retroperitoneal soft tissue sarcomas, who underwent postoperative radiotherapy between 1985 and 2003, were analyzed. The median follow-up period was 77 months (range, $8{\sim}240$ months). A total of 21 patients presented with primary disease, and two patients presented with recurrent disease. Liposarcomas and leiomyosarcomas represented 78% of the diagnosed tumor cases. Moreover, 17 cases were of high grade (grade 2 or 3). The median tumor size was 13 cm (range, $3{\sim}50\;cm$). Complete excision was achieved in 65% of patients. The median radiation dose was 50.4 Gy (range, 45.0 to 59.4 Gy), with conventional fractionation. Results: The 5-year overall, local recurrence-free, and distant metastasis-free survival rates were 68%, 58%, and 71%, respectively. Eleven patients experienced local recurrence, while 9 patients experienced distant metastasis. The most common site for distant metastasis was the liver. A univariate analysis revealed that adjacent organ invasion and age (>60 years) as the significant risk factors contributing to the prediction of poor overall survival. Moreover, multivariate analyses indicated that adjacent organ invasion remained significantly associated with a higher risk of death. In addition, patient age (>60 years) was the other identified risk factor for local recurrence by univariate and multivariate analyses. Except for one case of grade 3 diarrhea, no patient suffered grade 3 or higher complications. Conclusion: Our results were comparable to previous reports in that adjacent organ invasion and patient age (>60 years) were significant predictors of poor survival and tumor recurrence, respectively.

• Nuclear Medicine and Molecular Imaging
• /
• v.43 no.5
• /
• pp.429-435
• /
• 2009

Purpose: The purpose of this study was to assess the prognostic value of preoperative FDG-PET in colorectal cancer (CRC) patients with hepatic metastasis (HM). Materials and Methods: 24 CRC patients (M:F=14:10; age, $63{\pm}10$ yrs) with HM who had undergone preoperative FDG PET were included. Cure-intent surgery was performed in all the patients and HMs were controlled using resection (n=13), radio-frequency ablation (RFA) (n=7), and resection plus RFA (n=4). Potential prognostic markers tested were maxSUV of primary tumor, maxSUV of HM, maxSUV ratio of HM over primary tumor (M/P ratio), histologic grade, CEA level, venous/lymphatic/nerve invasion, T stage, N stage, no. of HM, no. of lymph node metastasis, and treatment modality of HM. Results: 14 CRC patients developed a recurrence with a median follow-up duration of 244 days, whereas 10 patients did not develop recurrence with a median follow-up duration of 504 days. M/P ratios but other potential prognostic markers were significantly higher in the recurrent patients ($0.72{\pm}0.14$) than recurrence-free patients ($0.54{\pm}0.23$) (p=0.038). M/P ratio only was found to predict recurrence by Cox multivariate analysis (hazard ratio 37.7, 95% confidence interval 2.01-706.1, p=0.016). The 11 patients with lower M/P ratio of <0.61 had significantly better disease-free survival rate than the 13 patients with higher M/P ratio (${\geq}0.61$) (p=0.026). Conclusion: maxSUV ratio of HM over primary tumor (M/P ratio) may be useful for prognosis prediction of CRC patients with HM. Higher FDG uptake of HM than that of primary tumor may indicate a more advanced status in stage IV CRC.

• Asian Pacific Journal of Cancer Prevention
• /
• v.17 no.9
• /
• pp.4483-4486
• /
• 2016

Background: Despite the fact that ovarian cancer is the seventh most common cancer in women worldwide and the fifth leading cause of cancer death, It is the most common cause of death due to reproductive cancers in Thailand where epithelial ovarian cancer (EOC) is commonly found. According to a Thai statistical analysis in 2010 by the Department of Medical Services, epithelial ovarian cancer was the sixth most common cancer in Thailand from 2001to 2003.The incidence of 5.1 per 100,000 women per year. Human epididymis protein 4 (HE4) is a novo diagnostic tumor marker for EOC. The combination of HE4 and carcinoma antigen 125 (CA 125) is a tool for detecting epithelial ovarian cancer (EOC) better than using CA 125 alone. Therefore, the researcher is interested in HE4 does have a role to predict recurrent epithelial ovarian cancer. Materials and Methods: The patients who had complete response after diagnosed with epithelial ovarian cancer by pathology, FIGO stage 3 or more had been treated through surgery and chemotherapy at the Sunpasitthiprasong Hospital from June 2014 until March 2016. The patients were followed up every three months, using tumor marker (CA 125, HE4,Carcinoma antigen 19-9) together with other checkup methods, such as rectovaginal examination, CXR every year and other imaging as indication. Afterwards, the data was analyzed for the ability of HE4 to detect recurrence of epithelial ovarian cancer. Results: In 47 patients in this study follow-up for 22 months after complete response treatment from surgery and chemotherapy in epithelial ovarian cancer, 23 had recurrent disease and HE4 titer rising. The patients with recurrent epithelial ovarian cancer demonstrated high levels of both HE4 and CA125 with sensitivity of 91.3% and 52.7% respectively, specificity of 87.5% and 95.6% and positive predictive values of 87.5% and 85.7%. HE4 can predict recurrent epithelial ovarian cancer (p-value=0.02242). Comparing HE4 and CA125 in predicting recurrent epithelial ovarian cancer HE4 had more potential than CA125 (p-value =0.8314). Conclusions: The present study showed HE4 to have a role in predicting recurrent epithelial ovarian cancer and HE4 is potentially better than CA125 as a marker for this purpose.

• Journal of the Korea Computer Graphics Society
• /
• v.28 no.1
• /
• pp.1-9
• /
• 2022

Non-small cell lung cancer (NSCLC) accounts for a high proportion of 85% among all lung cancer and has a significantly higher mortality rate (22.7%) compared to other cancers. Therefore, it is very important to predict the prognosis after surgery in patients with non-small cell lung cancer. In this study, the types of preoperative chest CT image patches for non-small cell lung cancer patients with tumor as a region of interest are diversified into five types according to tumor-related information, and performance of single classifier model, ensemble classifier model with soft-voting method, and ensemble classifier model using 3 input channels for combination of three different patches using pre-trained ResNet and EfficientNet CNN networks are analyzed through misclassification cases and Grad-CAM visualization. As a result of the experiment, the ResNet152 single model and the EfficientNet-b7 single model trained on the peritumoral patch showed accuracy of 87.93% and 81.03%, respectively. In addition, ResNet152 ensemble model using the image, peritumoral, and shape-focused intratumoral patches which were placed in each input channels showed stable performance with an accuracy of 87.93%. Also, EfficientNet-b7 ensemble classifier model with soft-voting method using the image and peritumoral patches showed accuracy of 84.48%.

• Asian Pacific Journal of Cancer Prevention
• /
• v.17 no.5
• /
• pp.2655-2660
• /
• 2016

Background: Prostate cancer is the second most common male cancer and remains a leading cause of cancer death worldwide. Heterogeneity regarding recurrence, tumor progression and therapeutic response reflects the inadequacy of traditional prognostic factors and underlies interest in new genetic and molecular markers. In this work, we studied the prognostic value of the expression of 9 proteins, Ki-67, p53, Bcl-2, PSA, HER2, E-cadherin, $p21^{WAF1/Cip1}$, $p27^{Kip1}$ and $p16^{ink4a}$ in prostate cancer. Materials and Methods: We conducted a retrospective study of 50 prostate cancers diagnosed in Pathology Department of Farhet Hached Hospital, Sousse, Tunisia, during a period of 12 months. Clinico-pathological data and survival were investigated. Protein expression was analyzed by immunohistochemistry on archived material. Results: Expression or over-expression of Ki-67, p53, Bcl-2, PSA, HER2, E-Cadherin, $p21^{WAF1/Cip1}$, $p27^{Kip1}$ and $p16^{ink4a}$ was observed in 68%, 24%, 32%, 78%, 12%, 90%, 20%, 44% and 56% of cases, respectively. Overall five-year survival was 68%. A statistically significant correlation was observed between death occurrence and advanced age (p=0.018), degree of tumor differentiation (p=0.0001), perineural invasion (p=0.016) and metastasis occurrence (p=0.05). Death occurrence was significantly correlated with the expression of p53 (p=0.007), Bcl-2 (p=0.02), Ki-67 (p=0.05) and $p27^{Kip1}$ (p=0.04). Conclusions: The p53, Bcl-2, Ki-67 and $p27^{Kip1}$ proteins may be useful additional prognostic markers for prostate cancer. The use of these proteins in clinical practice can improve prognosis prediction, disease screening and treatment response of prostatic cancer.

• Korean Journal of Head & Neck Oncology
• /
• v.31 no.2
• /
• pp.16-20
• /
• 2015

Background and Objectives : In papillary thyroid carcinoma (PTC), multiplicity and central or lateral lymph node metastases significantly affect the recurrence. This study was carried out to evaluate the clinical and histological characteristics of PTC according to the tumor size. Materials and Method : Between January 1, 2009 and December 31, 2014, 12,269 PTC patients underwent thyroid surgery at the Thyroid Cancer Center, Gangnam Severance Hospital, Yonsei University College of Medicine, Korea. We analyzed pathologic findings and clinical features according to the size of tumor Results : The mean size of tumor was $0.89{\pm}0.70cm$. The Central and lateral compartment metastases were observed 64.7% and 37.6% on the range that the primary tumor size is 1cm. There was a significant association between the PTC primary tumor size and multiplicity and cervical neck metastasis (p<0.001). Conclusion : The PTC primary tumor size for prediction of multiplicity and neck node metastasis can be helpful in optimization of the surgical extent for each patient.