• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• 제29권2호
• /
• pp.116-122
• /
• 2003

Pleomorphic adenoma is the most common salivary neoplasm mainly occurring in the major salivary glands - especially in parotid gland, which is characterized by variable histopathologic appearances and high recurrence rate with malignant transformation according to surgical situations. And this benign mixed tumor occurring in minor salivary glands is believed to shows same clinicopathologic appearances and relatively low recurrent rate compared with the case in major salivary glands. But there are few comparative studies of large series of pleomorphic adenoma occurring in minor salivary glands which includes different histopathologic appearance, clinical characteristics, treatment methods, recurrence rate, and malignant transformation. We retrospectively studied the 54 patients who were pathologically confirmed with pleomorphic adenoma occurring in minor salivary glands, and analyzed the clinico-histopathological appearance, surgical methods, recurrent cases. The results obtained are as follows. 1. The incidence of the tumor was most frequent in 4th & 5th decade, and in female. 2. Palate(90%) including hard & soft palate was the most frequent site for pleomorphic adenoma in minor salivary glands. 3. The exact duration could not be known due to asymptomatic slow growth patterns of the tumor. 4. The mean tumor size was 2.3cm. 5. 28 (52%) pleomorphic adenomas were classified as Cellular type (cell-rich), 17 (31%) specimen as Intermediate type(equal cell to stroma ratio), and 9 (17%) as Myxoid type(stroma-rich). 6. Surgically 51 cases (94%) were showed well-encapsulated tumors, but histopathologically only 34 specimen (63%) were wellencapsulated. Therefore pleomorphic adenomas in minor salivary glands also have to be excised more widely, not enucleated. And in case of suspicious malignancy or large tumor, preoperative incisional biopsy can be applied in the center of the tumor for prevention of rupture of tumor cell, and total excision with use of frozen biopsy for detection of malignancy and confirming the excision margin, and closed follow-up according to final histopathologic results is recommended.

• 사회복지연구
• /
• 제44권3호
• /
• pp.323-346
• /
• 2013

본 연구의 목적은 근로빈곤층이 경험한 빈곤이력에 따른 노동경력을 추적하여 총체적으로 파악하는데 있다. 이를 위해 한국노동패널 1-10차 년도의 결합자료를 이용하여 사건배열분석을 실시하였다. 분석의 결과, 일시빈곤층은 상대적으로 안정적인 노동경력을 공유하는데 반해, 반복빈곤층과 지속빈곤층은 불안정 저소득 저숙련의 일자리를 순환하는 특성을 보이며, 고용과 빈곤 간 상호연관성 기제가 약화되어 있다. 또한 근로빈곤층의 노동경력은 주로 상용직을 경험하는 유형, 임시 일용직을 경험하는 유형, 자영업을 지속하는 유형, 일자리 전환이 활발하다 상용직을 경험하는 유형과 자영업을 경험하는 유형 등의 5개 유형으로 구분된다. 빈곤이력에 따른 노동경력 유형의 상이한 경로가 관찰되며, 하위집단별 차별화된 정책적 개입이 요구된다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2023년도 학술발표회
• /
• pp.246-246
• /
• 2023

Multi-purpose dams are operated accounting for both physical and socioeconomic factors. This study aims to evaluate the utility of a deep learning algorithm-based model for three multi-purpose dam operation (Seomjin River dam, Juam dam, and Juam Control dam) in Seomjin River. In this study, the Gated Recurrent Unit (GRU) algorithm is applied to predict hourly water level of the dam reservoirs over 2002-2021. The hyper-parameters are optimized by the Bayesian optimization algorithm to enhance the prediction skill of the GRU model. The GRU models are set by the following cases: single dam input - single dam output (S-S), multi-dam input - single dam output (M-S), and multi-dam input - multi-dam output (M-M). Results show that the S-S cases with the local dam information have the highest accuracy above 0.8 of NSE. Results from the M-S and M-M model cases confirm that upstream dam information can bring important information for downstream dam operation prediction. The S-S models are simulated with altered outflows (-40% to +40%) to generate the simulated water level of the dam reservoir as alternative dam operational scenarios. The alternative S-S model simulations show physically inconsistent results, indicating that our deep learning algorithm-based model is not explainable for multi-purpose dam operation patterns. To better understand this limitation, we further analyze the relationship between observed water level and outflow of each dam. Results show that complexity in outflow-water level relationship causes the limited predictability of the GRU algorithm-based model. This study highlights the importance of socioeconomic factors from hidden multi-purpose dam operation processes on not only physical processes-based modeling but also aritificial intelligence modeling.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제12권8호
• /
• pp.355-364
• /
• 2023

최근 지능화된 사이버 위협이 지속적으로 증가함에 따라 기존의 패턴 혹은 시그니처 기반의 침입 탐지 방식은 새로운 유형의 사이버 공격을 탐지하는데 어려움이 있다. 따라서 데이터 학습 기반 인공지능 기술을 적용한 이상 징후 탐지 방법에 관한 연구가 증가하고 있다. 또한 지도학습 기반 이상 탐지 방식은 학습을 위해 레이블 된 이용 가능한 충분한 데이터를 필요로 하기 때문에 실제 환경에서 사용하기에는 어려움이 있다. 최근에는 정상 데이터로 학습하고 데이터 자체에서 패턴을 찾아 이상 징후를 탐지하는 비지도 학습 기반의 방법에 대한 연구가 활발히 진행되고 있다. 그러므로 본 연구는 시퀀스 로그 데이터로부터 유용한 시퀀스 정보를 보존하는 잠재 벡터(Latent Vector)를 추출하고, 추출된 잠재 벡터를 사용하여 이상 탐지 학습 모델을 개발하는데 있다. 각 시퀀스의 특성들에 대응하는 밀집 벡터 표현을 생성하기 위하여 Word2Vec을 사용하였으며, 밀집 벡터로 표현된 시퀀스 데이터로부터 잠재 벡터를 추출하기 위하여 비지도 방식의 오토인코더(Autoencoder)를 사용하였다. 개발된 오토인코더 모델은 시퀀스 데이터에 적합한 순환신경망 GRU(Gated Recurrent Unit) 기반의 잡음 제거 오토인코더, GRU 네트워크의 제한적인 단기 기억문제를 해결하기 위한 1차원 합성곱 신경망 기반의 오토인코더 및 GRU와 1차원 합성곱을 결합한 오토인코더를 사용하였다. 실험에 사용된 데이터는 시계열 기반의 NGIDS(Next Generation IDS Dataset) 데이터이며, 실험 결과 GRU 기반의 오토인코더나, 1차원 합성곱 기반의 오토인코더를 사용한 모델보다 GRU와 1차원 합성곱을 결합한 오토인코더가 훈련 데이터로부터 유용한 잠재 패턴을 추출하기 위한 학습 시간적 측면에서 효율적이었고 이상 탐지 성능 변동의 폭이 더 작은 안정된 성능을 보였다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제13권6호
• /
• pp.2639-2642
• /
• 2012

Background: The patterns of gastric cancer recurrence vary across societies. We designed the current study in an attempt to evaluate and reveal the outbreak of the recurrence patterns of gastric cancer and also prediction of time to recurrence and its effected factors in Iran. Materials and Methods: This research was performed from March 2003 to February 2007. Demographic characteristics, clinical and pathological diagnosis and classification including pathologic stage, tumor grade, tumor site and tumor size in of patients with GC recurrent were collected from patients' data files. To evaluate of factors affected on the relapse of the GC patients, gender, age at diagnosis, treatment type and Hgb were included in the research. Data were analyzed using Kaplan-Meier and logistic regression models. Results: After treatment, 82 patients suffered recurrence, 42, 33 and 17 by the ends of first, second and third years. The mean ( SD) and median ( IQR) time to recurrence in patients with GC were 25.5 (20.6-30.1) and 21.5 (15.6-27.1) months, respectively. The results of multivariate analysis logistic regression showed that only pathologic stage, tumor grade and tumor site significantly affected the recurrence. Conclusions: We found that pathologic stage, tumor grade and tumor site significantly affect on the recurrence of GC which has a high positive prognostic value and might be functional for better follow-up and selecting the patients at risk. We also showed time to recurrence to be an important factor for follow-up of patients.

• 산업경영시스템학회지
• /
• 제46권3호
• /
• pp.123-138
• /
• 2023

The Fourth Industrial Revolution and sensor technology have led to increased utilization of sensor data. In our modern society, data complexity is rising, and the extraction of valuable information has become crucial with the rapid changes in information technology (IT). Recurrent neural networks (RNN) and long short-term memory (LSTM) models have shown remarkable performance in natural language processing (NLP) and time series prediction. Consequently, there is a strong expectation that models excelling in NLP will also excel in time series prediction. However, current research on Transformer models for time series prediction remains limited. Traditional RNN and LSTM models have demonstrated superior performance compared to Transformers in big data analysis. Nevertheless, with continuous advancements in Transformer models, such as GPT-2 (Generative Pre-trained Transformer 2) and ProphetNet, they have gained attention in the field of time series prediction. This study aims to evaluate the classification performance and interval prediction of remaining useful life (RUL) using an advanced Transformer model. The performance of each model will be utilized to establish a health index (HI) for cutting blades, enabling real-time monitoring of machine health. The results are expected to provide valuable insights for machine monitoring, evaluation, and management, confirming the effectiveness of advanced Transformer models in time series analysis when applied in industrial settings.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제6권9호
• /
• pp.457-464
• /
• 2017

실내 공기 품질 측정은 측정 대상 공간의 대기 상태 유지, 외부 변인으로 인한 대기 이상 현상을 검출하려는 방법이다. 실내 공기 품질을 주기적으로 기록하면서 변인에 따른 공기 변화에 특정 패턴이 발생함을 관측할 수 있었으나, 파라미터를 설정하고 계수를 찾아 나가기엔 파라미터의 개수나 그 영향력을 추산하기 어렵고 결과가 시간에 의존적이라는 문제가 있다. 따라서 본 실험은 이것을 공식화하는 대신, 측정 주기마다 추이를 예측하는 관측치 중심의 기계 학습 모델을 개발하는 것을 목표로 한다. 본 논문은 실내 대기 품질을 주기적으로 전송 및 저장하는 측정기의 기록 데이터로 공기 품질 변화를 예측하는 모델을 설명하고 시계열 분석 모델을 구축한다.

• 대한두경부종양학회지
• /
• 제33권1호
• /
• pp.31-34
• /
• 2017

First branchial cleft anomaly is a very rare disease and exhibits various clinical presentations. Therefore, the diagnosis of first branchial cleft anomaly may be difficult; the condition is often misdiagnosed and mismanaged. Accurate diagnosis is very important, because if not diagnosed correctly, patients with first branchial cleft anomaly would be treated with local incision and drainage repeatedly. We report two cases of first branchial cleft anomaly. The first patient visited for recurrent swell and discharge in the infra-auricular area with a history of previous incision and drainage. The other patient showed a cystic mass in the infra-auricular area and all of them were misdiagnosed initially by their treating specialists elsewhere. The objective of this study is to share our experiences of first branchial cleft anomaly, and emphasize its various clinical patterns and the significance of accurate diagnosis.

• Animal Systematics, Evolution and Diversity
• /
• 제32권2호
• /
• pp.105-111
• /
• 2016

The genetic structure of marine animals that inhabit the seashore is affected by numerous factors. Of these, gene flow and natural selection during recruitment have strong influences on the genetic structure of seashore-dwelling species that have larval periods. Relative contributions of these two factors to the genetic structure of marine species would be determined mainly by the duration of larval stage. The relationship between larval period and genetic structure of population has been rarely studied in Korea. In this study, genetic variations of cytochrome oxidase subunit I (COI) were analyzed in two dominant species on rocky shore habitats in the Korean peninsula: periwinkle Littorina brevicula and acorn barnacle Fistulobalanus albicostatus. Both species are not strongly structured and may have experienced recent population expansion. Unlike periwinkle, however, barnacle populations have considerable genetic variation, and show a bimodal pattern of mismatch distribution. These results suggest that barnacle populations are more affected by local adaptation rather than gene flow via larval migration. The bimodal patterns of barnacle populations observed in mismatch distribution plots imply that they may have experienced secondary contact. Further studies on seashore-dwelling species are expected to be useful in understanding the evolution of the coastal ecosystem around Korean waters.

• BMB Reports
• /
• 제51권1호
• /
• pp.1-2
• /
• 2018

Positive selection on a new beneficial mutation generates a characteristic pattern of DNA sequence polymorphism when it reaches an intermediate allele frequency. On genome sequences of African Drosophila melanogaster, we detected such signatures of selection at 37 candidate loci and identified "sweeping haplotypes (SHs)" that are increasing or have increased rapidly in frequency due to hitchhiking. Based on geographic distribution of SH frequencies, we could infer whether selective sweeps occurred starting from de novo beneficial mutants under simple constant selective pressure. Single SHs were identified at more than half of loci. However, at many other loci, we observed multiple independent SHs, implying soft selective sweeps due to a high beneficial mutation rate or parallel evolution across space. Interestingly, SH frequencies were intermediate across multiple populations at about a quarter of the loci despite relatively low migration rates inferred between African populations. This invokes a certain form of frequency-dependent selection such as heterozygote advantage. At one locus, we observed a complex pattern of multiple independent that was compatible with recurrent frequency-dependent positive selection on new variants. In conclusion, genomic patterns of positive selection are very diverse, with equal contributions of hard and soft sweeps and a surprisingly large proportion of frequency-dependent selection in D. melanogaster populations.