• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.5
• /
• pp.197-202
• /
• 2022

IoT (Internet of Things) devices are being attacked by malware due to many security vulnerabilities, such as the use of weak IDs/passwords and unauthenticated firmware updates. However, due to the diversity of CPU architectures, it is difficult to set up a malware analysis environment and design features. In this paper, we design time series features using the byte sequence of executable files to represent independent features of CPU architectures, and analyze them using recurrent neural networks. The proposed feature is a fixed-length time series pattern extracted from the byte sequence by calculating partial entropy and applying linear interpolation. Temporary changes in the extracted feature are analyzed by RNN and LSTM. In the experiment, the IoT malware detection showed high performance, while low performance was analyzed in the malware family classification. When the entropy patterns for each malware family were compared visually, the Tsunami and Gafgyt families showed similar patterns, resulting in low performance. LSTM is more suitable than RNN for learning temporal changes in the proposed malware features.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.10a
• /
• pp.535-538
• /
• 2001

Since 1990s, many literatures have shown that connectionist models, such as back propagation, recurrent network, and RBF (Radial Basis Function) outperform the traditional models, MA (Moving Average), AR (Auto Regressive), and ARIMA (Auto Regressive Integrated Moving Average) in time series prediction. Neural based approaches to time series prediction require the enough length of historical measurements to generate the enough number of training patterns. The more training patterns, the better the generalization of MLP is. The researches about the schemes of generating artificial training patterns and adding to the original ones have been progressed and gave me the motivation of developing VTG schemes in 1996. Virtual term is an estimated measurement, X(t+0.5) between X(t) and X(t+1), while the given measurements in the series are called actual terms. VTG (Virtual Tern Generation) is the process of estimating of X(t+0.5), and VTG schemes are the techniques for the estimation of virtual terms. In this paper, the alternative VTG schemes to the VTG schemes proposed in 1996 will be proposed and applied to multivariate time series prediction. The VTG schemes proposed in 1996 are called deterministic VTG schemes, while the alternative ones are called stochastic VTG schemes in this paper.

• Journal of Korean Neurosurgical Society
• /
• v.39 no.2
• /
• pp.144-147
• /
• 2006

Rhabdoid meningioma is an uncommon meningioma variant, which was initially described in 1998 by Perry et al. It is classified as a variant corresponding to grade III in the recent World Health Organization[WHO] classification of tumors of the nervous system, because of its unique histopathological features and its highly aggressive clinical course. To date there were still a lot of arguments about their developmental patterns and the treatment strategy especially for rhabdoid meningiomas lacking other histological features of malignancy. Although the exact causes of rhabdoid transformation of a benign meningioma are unknown, malignant transformation long duration after surgical resection under an underlying predilection for malignancy is suggested in our case. And we suggest that any histological rhabdoid features in recurrent meningiomas or even in primary cases seem to indicate the malignant clinical course and the need for aggressive treatments, because transformation from a benign or atypical one to a malignant one seems to occur at last.

• Korean Journal of Head & Neck Oncology
• /
• v.28 no.1
• /
• pp.27-30
• /
• 2012

Basaloid squamous cell carcinoma(BSCC) is a high-grade variant of squamous cell carcinoma, with a prediction for multifocal involvement of the base of tongue, pyriform sinus, supraglottic larynx, hypopharynx and palatine tonsil. It primary affects men in the seventh decade of life with frequent cervical lymph-node metastasis at presentation. Grossly, these tumors are usually firm to hard, with associated central necrosis, occuring as exophytic to nodular masses. Histologically, the this infiltrating tumor offers a variety of growth patterns, including solid, lobular, cribriform, cords, trabeculae, nests and glands or cyst. We present a 55-year-old female who was treated with surgical excision and radiotherapy. She was firstly presented as a recurrent inflammatory neck mass and finally diagnosed with basaloid squamous cell carcinoma in the palatine tonsil.

• Women's Health Nursing
• /
• v.28 no.2
• /
• pp.145-153
• /
• 2022

Purpose: This study aimed to investigate the influencing factors of breast cancer recurrence by comparing the risk factors and lifestyle patterns related to breast cancer in Korean women with and without recurrence. Methods: This cross-sectional survey comprised 241 Korean women diagnosed with breast cancer who had received follow-up treatment. Participants were recruited from a university hospital in Seoul and an online social media platform for breast cancer patients. Data were collected either via online or a paper survey, using a structured questionnaire that included general and disease-related characteristics and lifestyle behaviors. Data were analyzed using descriptive statistics, univariate analysis, and logistic regression. Results: Recurrence of breast cancer was influenced by four factors; childbirth experience, consumption of green/yellow vegetables, drinking behavior, and recovery from fatigue after sleep. Prevalence of recurrent breast cancer was associated with no childbirth experience (OR=2.29, p=.010), fewer green/yellow vegetables (OR=0.71, p=.008), drinking behavior (OR=0.24, p=.001), and a lower level of recovery from fatigue after sleep (OR=0.51, p<.001). Conclusion: Aside from having experienced childbirth, this study identified several modifiable factors that influence breast cancer recurrence. Increasing green/ yellow vegetable intake, alleviating fatigue, and reducing alcohol intake are important. Intervention strategies in clinical research and practice can be applied to address risk factors and reduce the prevalence of recurrent breast cancer.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.19
• /
• pp.8259-8263
• /
• 2014

Aims: To determine the clinical characteristics, pathological features, local and distant failure patterns in patients with carcinoma of major salivary glands treated with surgery and postoperative radiotherapy (PORT). Materials and Methods: We retrospectively reviewed 106 cases of major salivary gland tumor seen at our centre (1998-2008). Sixty five cases of major salivary gland carcinoma were selected for analysis (exclusions: benign, palliative, non-carcinomas). The patient population treated by surgery and PORT was divided into two groups: 1) Patients who underwent surgery and immediate PORT (Primary PORT); 2) Patients with recurrent carcinoma who underwent at least two surgeries and received PORT in the immediate post-operative period of the last performed surgery (Recurrent PORT). Recurrence free survival (RFS) was assessed using the Kaplan-Meier method. Results: Median age was 35 years with a male: female ratio of 1.3:1. The majority of cancers were located in the parotid gland (86.2%) and the most common histology was mucoepidermoid carcinoma (43%). Thirty nine cases (60%) were primary while 26 (40%) were recurrent. Optimal surgery was performed in 59/65 patients (90.8%). 43 patients (66.2%) underwent neck dissection, of which 14 (32.5%) had nodal metastasis. Overall, 61 (93.8%) patients complied with the prescribed radiotherapy. Median dose of PORT was 60 Gy. Median follow-up was 13.1 months (range 2-70). Relapse free survival was 50.4% at 60 months. Some 12 cases (18.5%) recurred with a median time to recurrence of 16.9 months. Conclusions: Surgery and PORT is an effective treatment for major salivary gland carcinoma with over 90% compliance and <20% recurrence. Early treatment with postoperative radiotherapy may increase the survival rate in major salivary gland carcinoma patients.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.10
• /
• pp.4393-4399
• /
• 2015

Background: To explore the independent prognostic factors for the recurrence/metastasis of patients with locoregionally advanced nasopharyngeal carcinoma (LANPC). Materials and Methods: A total of 604 patients initially diagnosed as LANPC by pathohistology in Fujian Provincial Cancer Hospital were selected to analyze the relationship between the clinical pathological patterns, therapeutic protocols and clinical stages with the recurrence/metastasis of LANPC. Results: The 1-, 3- and 5-year locoregionally recurrent rates of LANPC patients were 2.0%, 9.5% and 12.9% respectively, with average recurrent period being 78 months. Univariate analysis results indicated that clinical stages had certain influence on the recurrent period of LANPC patients. However, COX regression models showed that ages, genders and clinical stages were not the independent prognostic factors influencing the recurrence. The 1-, 3- and 5-year metastatic rates of LANPC patients were 6.6%, 17.5% and 18.8% respectively, with average metastatic period of 73 months. Univariate analysis results demonstrated that ages, N stages, clinical stages, locations of lymph node, retropharyngeal lymph node and extracapsular invasion of lymph node had certain influence on the metastatic period of LANPC patients. Additionally, further COX regression analysis results suggested that T stages, reduction protocols and extracapsular invasion of lymph node were the independent prognostic factors influencing the metastasis of patients with LANPC, in which T stages and extracapsular invasion of lymph node were the pestilent factors while reduction protocols the protective factor. Conclusions: Induction chemotherapy is beneficial to LANPC patients with initial treatment, and the metastatic rate decreases greatly after the application of reduction chemotherapy.

• Pediatric Infection and Vaccine
• /
• v.25 no.2
• /
• pp.107-112
• /
• 2018

Staphylococcus aureus is now a major community-acquired pathogen worldwide, notably associated with skin and soft tissue infections. Staphylococci are present in the form of colonizers or environmental contaminants at home and increase the risk of recurrent infection. We are describing recurrent familial furunculosis caused by Panton-Valentine Leukocidin-positive methicillin susceptible S. aureus ST1 in Korea. An infant, his father and mother had furunculosis due to methicillin-sensitive S. aureus (MSSA) infection with identical susceptibility patterns. ST1 accounted for all 3 isolates and they were confirmed of having agr group I. Both sec and seh were detected in all isolates using polymerase chain reaction (PCR) assays, and all isolates contained Panton-Valentine leukocidin (PVL) genes. Risk factors for the household spread of S. aureus include skin conditions and close physical contact among household members. The relationship between S. aureus colonization of household contacts and the occurrence of S. aureus infection should be studied into more detail.

• Radiation Oncology Journal
• /
• v.21 no.2
• /
• pp.143-148
• /
• 2003

Purpose: The purpose of this study was to evaluate treatment results in terms of the survival and failure patterns subsequent to radiation therapy in recurrent cervical cancer, fellowing primary surgery. Material and Methods: Between January 1990 and December 1999, 27 patients, with recurrent cervical cancer following primary surgery, were subsequently treated with radiation in the Department of Radiation Oncology, at the Keimyung University Dongsan Medical Center. Their median age was 48, ranging from 31 to 70 years old. With regard to the Initial FIGO stage on presentation, 20 and 7 patients were stages I and II, respectively. Twenty three patients had squamous cell carcinomas and 4 had adenocarcinomas. The time interval from the primary surgery to the recurrence ranged from 2 to 90 months with a median of 29 months. The recurrent sites were the vaginal cuff alone, the pelvic cavity and combined recurrence in 14, 9 and 4 patients, respectively. Radiation was peformed, with external and vaginal intracavitary radiation in 13 patients, external radiation alone in 13 and vaginal intracavitary radiation alone in another one. The median follow-up period was 55 months, ranging from 6 to 128 months. Results: The five year disease free survival (5y DFS) and five year overall survival (5y OS) rates were 68.2 and 71.9$\%$, respectively. There was a marginal statistically significant difference in the 5y DFS in relation to the recurrent site (5y DFS, 85.7$\%$ in vaginal cuff recurrence alone, 53.3$\%$ in pelvic cavity recurrence, p=0.09). There was no difference in the survival according to the time interval between the primary surgery and a recurrence. There was only a 7$\%$ local failure rate in the patients with a vaginal cuff recurrence. The major failure patterns were local failure in the patients with pelvic cavity recurrence, and distant failure in the patients with a combined recurrence. There were no complications above grade 3 after the radiation therapy. Conclusion: Radiation therapy was safe and effective treatment for a recurrent carcinoma of the uterine cervix following primary surgery, especially the external beam radiation and vaginal intracavitary irradiation achieved the best results in the patients with a vaginal cuff recurrence following primary surgery.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.22 no.3
• /
• pp.239-248
• /
• 2020

Exact rock classification helps suitable support patterns to be installed. Face mapping is usually conducted to classify the rock mass using RMR (Rock Mass Ration) or Q values. There have been several attempts to predict the grade of rock mass using mechanical data of jumbo drills or probe drills and photographs of excavation surfaces by using deep learning. However, they took long time, or had a limitation that it is impossible to grasp the rock grade in ahead of the tunnel surface. In this study, a method to predict the Q value ahead of excavation surface is developed using recurrent neural network (RNN) technique and it is compared with the Q values from face mapping for verification. Among Q values from over 4,600 tunnel faces, 70% of data was used for learning, and the rests were used for verification. Repeated learnings were performed in different number of learning and number of previous excavation surfaces utilized for learning. The coincidence between the predicted and actual Q values was compared with the root mean square error (RMSE). RMSE value from 600 times repeated learning with 2 prior excavation faces gives a lowest values. The results from this study can vary with the input data sets, the results can help to understand how the past ground conditions affect the future ground conditions and to predict the Q value ahead of the tunnel excavation face.