• 한국산학기술학회논문지
• /
• 제21권6호
• /
• pp.374-382
• /
• 2020

본 연구에서 저장탄약 신뢰성평가(ASRP: Ammunition Stockpile Reliability Program)의 데이터 특성을 고려하여 입력변수를 줄이는 정규화기법을 제안함으로써 분류성능의 저하 없이 저장탄약 신뢰성분류 인경신경망모델의 학습 속도향상을 목표로 하였다. 탄약의 성능에 대한 기준은 국방규격(KDS: Korea Defense Specification)과 저장탄약 시험절차서(ASTP: Ammunition Stockpile reliability Test Procedure)에 규정되어 있으며, 평가결과 데이터는 이산형과 연속형 데이터가 복합적으로 구성되어 있다. 이러한 저장탄약 신뢰성평가의 데이터 특성을 고려하여 입력변수는 로트 추정 불량률(estimated lot percent nonconforming) 또는 고장률로 정규화 하였다. 또한 입력변수의 unitary hypercube를 유지하기 위하여 최소-최대 정규화를 2차로 수행하는 2단계 정규화 기법을 제안하였다. 제안된 2단계 정규화 기법은 저장탄약 신뢰성평가 데이터를 이용하여 비교한 결과 최소-최대 정규화와 유사하게 AUC(Area Under the ROC Curve)는 0.95 이상이었으며 학습속도는 학습 데이터 수와 은닉 계층의 노드 수에 따라 1.74 ~ 1.99 배 향상되었다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제13권1호
• /
• pp.199-203
• /
• 2012

Purpose: To evaluate the prognostic value of serum CYFRA21-1, CEA and hemoglobin levels regarding long-term survival of patients with esophageal squamous cell carcinoma (ESCC) treated with concurrent chemoradiotherapy (CRT). Methods: Age, gender, Karnofsky Performance Status (KPS), tumor location, tumor length, T stage, N stage and serum hemoglobin, and CYFRA21-1 and CEA levels before concurrent CRT were retrospectively investigated and related to outcome in 113 patients receiving 5-fluorouracil and cisplatin combined with radiotherapy for ESCC. The Kaplan-Meier method was used to analyze prognosis, the log-rank to compare groups, the Cox proportional hazards model for multivariate analysis, and ROC curve analysis for assessment of predictive performance of biologic markers. Results: The median survival time was 20.1 months and the 1-, 2-, 3-, 5- year overall survival rates were 66.4%, 43.4%, 31.9% and 15.0%, respectively. Univariate analysis showed that factors associated with prognosis were KPS, tumor length, T-stage, N-stage, hemoglobin, CYFRA21-1 and CEA level. Multivariate analysis showed T-stage, N-stage, hemoglobin, CYFRA21-1 and CEA level were independent predictors of prognosis. By ROC curve, CYFRA21-1 and hemoglobin showed better predictive performance for OS than CEA (AUC= 0.791, 0.704, 0.545; P=0.000, 0.000, 0.409). Conclusions: Of all clinicopathological and molecular factors, T stage, N stage, hemoglobin, CYFRA21-1 and CEA level were independent predictors of prognosis for patients with ESCC treated with concurrent CRT. Among biomarkers, CYFRA21-1 and hemoglobin may have a better predictive potential than CEA for long-term outcomes.

• 융합정보논문지
• /
• 제7권1호
• /
• pp.55-60
• /
• 2017

사물인터넷 환경에서 인간의 개입 없는 지능화된 서비스를 위해서는 IoT 디바이스에서 생성되는 빅데이터로 부터 정상 패턴을 학습하고 이를 기반으로 불량, 오작동과 같은 이상 징후에 대해 예측하는 과정이 요구된다. 본 연구의 목적은 제품 공정의 다양한 기기에서 발생되는 빅데이터를 분석함으로써 제품 불량을 예측할 수 있는 기계 학습모델을 구현하는 것이다. 기계 학습 모델은 어느 정도 볼륨을 가진 기존 데이터를 기반으로 분석을 해야 하므로 빅데이터 분석도구 R을 사용하였으며, 제품 공정에서 수집된 데이터에는 제품에 대한 불량 여부가 포함되어 있으므로 지도 학습 모델을 활용하였다. 연구의 결과, 제품 불량에 영향을 주는 변수 및 변수 조건을 분류하였고, 의사결정 트리를 기반으로 제품의 불량 여부에 대한 예측 모델을 제시하였다. 또한, ROC Curve를 이용한 모델의 적합성 및 성능평가 분석에서 모델의 예측력은 상당히 높게 나타났다.

• Physical Therapy Rehabilitation Science
• /
• 제7권3호
• /
• pp.102-108
• /
• 2018

Objective: To investigate whether the Minimal Clinically Important Difference (MCID) clinically defines improvement of Berg Balance Scale (BBS) scores in people with acute stroke in response to rehabilitation. Design: Retrospective study. Methods: Seventy-three participants with acute stroke participated in the study. Balance evaluation was performed using the BBS. All patients received rehabilitation with physical therapy for 4 weeks, 5 times a week, for 2 hours and 20 minutes a day. An anchor-based approach using the clinical global impression was used to determine the MCID of the BBS. The MCID was used to define the minimum change in the BBS total score (postintervention-preintervention) that was needed to perceive at least a 3-point improvement on the global rating of change. Receiver operating characteristic (ROC) curves was used to define the cut-off values of the optimal MCID of the BBS in order to discriminate between improvement and no improvement groups. Results: The optimal MCID cut-off point for the BBS change scores was 12.5 points for males with a sensitivity (Sn) of 0.62 and a specificity (Sp) of 0.89, and 12.5 points for females with a Sn of 0.69 and Sp of 0.85. The area under the curve of the ROC curve for all participants were 0.84 (95% confidence interval [CI], 0.72; 0.95, p<0.001), and 0.89 (95% CI, 0.77; 1.00, p<0.001), respectively. Conclusions: The MCID for improvement in balance as measured by the BBS was 13.5 points, indicating that the MCID does clinically detect changes in balance abilities in persons with stroke.

• 한국건강관리협회지
• /
• 제4권1호
• /
• pp.1-11
• /
• 2006

"본 논문은 대한외과학회지 2006년 제70권제1호에 실렸던 논문으로 대한외과학회 편집위원회 승인을 득하고 본 협회지에 게재함. Purpose: Malnutrition has been frequently reported for patients on their admission to the hospital and it has been associated with an increase in morbidity, mortality and the length of the hospital stay. Although a number of screening tools have been developed to identify those patients at risk for malnutrition, there is no' gold standard' for defining malnutrition and the malnourished patients remain largely unrecognized. The aim of this study is to evaluate the efficacy of a nutritional screening tool for use in Dankook University Hospital. Methods Nutritional evaluation was performed for 53 patients who were admitted to the department of surgery and internal medicine between October and December 2004. The screening tool was completed by the ward nurse and the nutritional support team nurse on the same patients within24 hours of admission. The nutritional support team nurse performed the full assessment. The screening sheet included 4 questions regarding body mass index, recent unintentional weight loss, food intake and disease severity. Each answer was scored and a total of 5 was tested as the criterion fey malnutrition. The full assessment included current body weight, recent weight loss, triceps skinfold thickness, mid-arm muscle circumference, serum albumin)in and total lymphocyte count. Malnutrition was defined by 3 or more values below the reference values. The reliability of the screening tool was assessed using kappa statistic. Sensitivity, specificity and accuracy were calculated to evaluate the validity of the screening tool. The receiver operating characteristic(ROC) curve was drawn to choose a cutoff valve that maximizes sensitivity and specificity. Results' The level of agreement between the ward nurse and the NST nurse was good for BMI and food intake and moderate for weight loss and disease severity. The full assessment identified7 patients(13.2%) as malnourished. The screening sheet had a sensitivity of 86% and a specificity of 80%. According to the ROC curve, a score of 5 points provided the best validity. Conclusion The nutritional screening tool is reliable when completed by different observers and it is valid for nutritional assessment.

• 생물정신의학
• /
• 제12권2호
• /
• pp.165-172
• /
• 2005

Objectives:The purpose of this study is to 1) conduct a discrimination analysis of schizophrenia and bipolar affective disorder using MMPI profile through artificial neural network analysis and logistic regression analysis, 2) to make a comparison between advantages and disadvantages of the two methods, and 3) to demonstrate the usefulness of artificial neural network analysis of psychiatric data. Procedure:The MMPI profiles for 181 schizophrenia and bipolar affective disorder patients were selected. Of these profiles, 50 were randomly placed in the learning group and the remaining 131 were placed in the validation group. The artificial neural network was trained using the profiles of the learning group and the 131 profiles of the validation group were analyzed. A logistic regression analysis was then conducted in a similar manner. The results of the two analyses were compared and contrasted using sensitivity, specificity, ROC curves, and kappa index. Results:Logistic regression analysis and artificial neural network analysis both exhibited satisfactory discriminating ability at Kappa index of greater than 0.4. The comparison of the two methods revealed artificial neural network analysis is superior to logistic regression analysis in its discriminating capacity, displaying higher values of Kappa index, specificity, and AUC(Area Under the Curve) of ROC curve than those of logistic regression analysis. Conclusion:Artificial neural network analysis is a new tool whose frequency of use has been increasing for its superiority in nonlinear applications. However, it does possess insufficiencies such as difficulties in understanding the relationship between dependent and independent variables. Nevertheless, when used in conjunction with other analysis tools which supplement it, such as the logistic regression analysis, it may serve as a powerful tool for psychiatric data analysis.

• 대한임상독성학회지
• /
• 제1권1호
• /
• pp.27-33
• /
• 2003

Purpose: According as the accessibility about drugs becomes various, the occurrence of drug intoxication is increasing. Since report that doxylamine causes rhabdomyolysis often, drug-induced rhabdomyolysis is one of the most important complications in patients with drug intoxication. Acute renal failure (ARF)'s availability is important to the management in rhabdomyolysis, but report about rhabdomyolysis or ARF occurrence for whole intoxicated drugs is lacking up to now. Methods: This research did to 61 patient who had rhabdomyolysis of drug intoxication. First, object patients were divided into two gruops: doxylamine-ingested (Group I) vs non-doxylamine ingested (Group II). And then we analyzed on the early patient's clinical events and laboratory data. We used ROC curve to recognize'the early clinical factors that could forecast ARF appearance among these patients in addition. Results: Almost rhabdomyolysis was happened by doxylamine in drug intoxication ($55.7\%$). However, as compared to group II, group I showed better clinical course, lesser ARF occurrence and hemodialysis requirement. In group II, time was longer in hospital reaching from intoxication, the ARF occurrence rate was higher ($52.6\%$). Analyzing the ROC curve to useful initial factors, they were creatinine, uric acid and interval time from ingestion to hospital. These cut-off values were 1.44 mg/dL, 6.8 mg/dL and 5 hrs. Sensitivity for ARF estimate was $100\%$, specificity $69-98\%$. Conclusion: Compared to group II, Doxylamine-ingested group showed good clinical course. Creatinine, uric acid, interval time from ingestion to hospital aided in ARF estimate in drug-induced rhabdomyolysis.

• 한국정보통신학회논문지
• /
• 제20권1호
• /
• pp.22-29
• /
• 2016

수중환경 하에서 표적을 탐지하고 식별하는 문제는 군사적인 목적은 물론 비군사적 목적으로도 많은 연구가 수행되어 왔다. 수중환경에서의 수중음향 신호가 시간 공간적으로 특성이 변화하며 천해 다중경로 환경을 반영하는 복잡한 특성을 보이는 점으로 인해 능동 표적인식 기술은 매우 어려운 기술로 여겨져 왔다. 본 논문에서는 Fractional 푸리에 변환의 기본 개념과 최적 변환 차수에 대해 설명하고, 이를 이용하여 LFM 신호의 시간-주파수 특성과 스펙트럼 사이의 관계를 분석한다. 그리고 이러한 분석결과를 바탕으로 능동소나 표적 탐지 기법을 제안한다. 제안된 방법의 성능을 검증하기 위해, 기존의 FFT를 이용한 정합필터와 성능을 비교하였다. AUC(Area Under the ROC Curve)의 측면에서 볼 때 제안된 방식이 기존의 방법보다 성능이 우수한 실험결과를 보였다.

• Clinical and Experimental Pediatrics
• /
• 제59권3호
• /
• pp.139-144
• /
• 2016

Purpose: Acute pyelonephritis (APN) is a serious bacterial infection that can cause renal scarring in children. Early identification of APN is critical to improve treatment outcomes. The neutrophil-lymphocyte ratio (NLR) is a prognostic marker of many diseases, but it has not yet been established in urinary tract infection (UTI). The aim of this study was to determine whether NLR is a useful marker to predict APN or vesicoureteral reflux (VUR). Methods: We retrospectively evaluated 298 pediatric patients ($age{\leq}36months$) with febrile UTI from January 2010 to December 2014. Conventional infection markers (white blood cell [WBC] count, erythrocyte sedimentation rate [ESR], C-reactive protein [CRP]), and NLR were measured. Results: WBC, CRP, ESR, and NLR were higher in APN than in lower UTI (P<0.001). Multiple logistic regression analyses showed that NLR was a predictive factor for positive dimercaptosuccinic acid (DMSA) defects (P<0.001). The area under the receiver operating characteristic (ROC) curve was high for NLR (P<0.001) as well as CRP (P<0.001) for prediction of DMSA defects. NLR showed the highest area under the ROC curve for diagnosis of VUR (P<0.001). Conclusion: NLR can be used as a diagnostic marker of APN with DMSA defect, showing better results than those of conventional markers for VUR prediction.

• Pediatric Gastroenterology, Hepatology & Nutrition
• /
• 제21권2호
• /
• pp.118-126
• /
• 2018

Purpose: Classical criteria for diagnosis of autoimmune hepatitis (AIH) are intended as research tool and are difficult to apply at patient's bedside. We aimed to study the accuracy of simplified criteria and the concordance with the expert diagnosis based on the original criteria. Methods: A cohort of children under study for liver disorder was selected through consecutive sampling to obtain the prevalence of AIH within the group of differential diagnoses. AIH was defined, based on classical criteria, through committee review of medical reports. Validity indicators of the simplified criteria were obtained in an intention to diagnose approach. Optimal cut-off and the area under the receiver operating characteristic (ROC) curve were calculated. Results: Out of 212 cases reviewed, 47.2% were AIH. For the optimal cut-off (6 points), the simplified criteria showed a sensitivity of 72.0% and a specificity of 96.4%, with a 94.7% positive and a 79.4% negative predictive value. The area under the ROC curve was 94.3%. There was a good agreement in the pre-treatment concordance between the classical and the simplified criteria (kappa index, 0.775). Conclusion: Simplified criteria provide a moderate sensitivity for the diagnosis of AIH, but may help in indicating treatment in cases under suspicion with 6 or more points.