• 전기전자학회논문지
• /
• 제20권1호
• /
• pp.82-88
• /
• 2016

영상 인식에서 널리 이용되는 PPHT(Progressive Probability Hough Transform)는 직선을 정확하게 인식하는 우수한 알고리즘이나 원본 영상이 선명하지 않거나 복잡하여 잡음 성분이 많은 경우 인식률이 감소하는 문제가 있다. 이러한 문제를 해결하기 위해 잡음에 강하고 손상된 가장자리 패턴을 복구하며 직선을 인식하는 개선된 PPHT 방식을 제안한다. 제안하는 알고리즘은 픽셀 단위로 직선을 추적하고 검증하여 선분을 검출하는 방식으로 잡음의 영향을 최소화하고 손상된 가장자리 패턴을 일정 범위 내에서 복구하여 인식률을 증가시켰다. 제안한 알고리즘을 차선 인식에 적용하여 직선의 오인식률을 30% 이상 감소시키고 선분 인식률이 15%까지 증가함을 확인하였다.

• Clinical and Experimental Pediatrics
• /
• 제58권5호
• /
• pp.183-189
• /
• 2015

Purpose: Catheter urine (CATH-U) and suprapubic aspiration (SPA) are reliable urine collection methods for confirming urinary tract infections (UTI) in infants. However, noninvasive and easily accessible collecting bag urine (CBU) is widely used, despite its high contamination rate. This study investigated the validity of CBU cultures for diagnosing UTIs, using CATH-U culture results as the gold standard. Methods: We retrospectively analyzed 210 infants, 2- to 24-month-old, who presented to a tertiary care hospital's pediatrics department between September 2008 and August 2013. We reviewed the results of CBU and CATH-U cultures from the same infants. Results: CBU results, relative to CATH-U culture results (${\geq}10^4$ colony-forming units [CFU]/mL) were widely variable, ranging from no growth to ${\geq}10^5CFU/mL$. A CBU cutoff value of ${\geq}10^5CFU/mL$ resulted in false-positive and false-negative rates of 18% and 24%, respectively. The probability of a UTI increased when the CBU bacterial count was ${\geq}10^5/mL$ for all infants, both uncircumcised male infants and female infants (likelihood ratios [LRs], 4.16, 4.11, and 4.11, respectively). UTIs could not be excluded for female infants with a CBU bacterial density of $10^4-10^5$ (LR, 1.40). The LRs for predicting UTIs based on a positive dipstick test and a positive urinalysis were 4.19 and 3.11, respectively. Conclusion: The validity of obtaining urine sample from a sterile bag remains questionable. Inconclusive culture results from CBU should be confirmed with a more reliable method.

• Asian Pacific Journal of Cancer Prevention
• /
• 제17권5호
• /
• pp.2605-2610
• /
• 2016

The International Myeloma Working Group considers the serum free light chain (SFLC) assay to be an adjunct to traditional tests. Apart from the FLC ratio, the absolute values of individual free light chains also are gaining importance as they appear to be more relevant in certain clinical settings. Automated assays are available for their determination. As laboratories put new test systems into use catering to different disease populations, they are required by accreditation and certification bodies to verify or establish performance specifications, including reference intervals (RIs) representative of their population. Our aim was to establish local RIs for SFLC in a multicentre representative healthy population using a robust method. There was no significant relationship between SFLC levels and age, gender and creatinine levels. The 95% RI for ${\kappa}SFLC$ was 4.81 to 33.86mg/L, for ${\lambda}$ SFLC was 5.19 to 23.67mg/L and for ${\kappa}/{\lambda}SFLC$ was 0.36 to 2.33, significantly higher than the values given by the manufacturer. The ${\kappa}/{\lambda}$ SFLC ratio at 2.23, covering 100% of the data, showed 72% sensitivity (95% CI=39.0 - 94.0), 100% specificity (95% CI=71.5 - 100.0), 100% PPV (95% CI=21.5 - 100.0), 95% NPV (95% CI=75.4 - 99.9), and 79% accuracy (95% CI=56.0 - 93.0). In the patient group, kit RI for ${\kappa}/{\lambda}$ SFLC ratio classified 45.5% (n=5) as positive vs 9.1% (n=1) positive by the study RI, while the kit RI for kappa FLC classified 90.9% (n=10) as positive vs 54.5% (n=6), indicating increased probability of false positive test results with the kit RI when applied to our patient population. Appropriate and specific reference intervals and criteria values result in fewer false-positive and false-negative results which means fewer wrong or missed diagnoses.

• Communications for Statistical Applications and Methods
• /
• 제18권4호
• /
• pp.445-455
• /
• 2011

표본추출에서 오분류된 이진자료는 흔히 발생될 수 있는 현실적인 문제이지만 통계적 방법론은 상대적으로 제한적이라고 할 수 있다. 특히, 모비율의 구간추정 문제는 고전적인 Wald 방법에 의존하고 있었다. 그러나 최근 이승천과 최병수 (2009)에서 Agresti-Coull 방법을 적용하고 새로운 구간추정 방법을 제시하였으며, 수치적인 방법에 의해 Agresti-Coull 신뢰구간의 효율성을 주장하였다. 본 연구에서는 오분류된 이진자료에 대한 베이지안 모형을 다루었으며, 베이지안 모형이 Agresti-Coull 신뢰구간의 이론적 배경이 될 수 있는지 살펴 보았다.

• 한국정보보호학회:학술대회논문집
• /
• 한국정보보호학회 2006년도 하계학술대회
• /
• pp.365-369
• /
• 2006

Multiple Message Authentication Codes can be represented by one of the Short MAC, Bloom Filter or Compressed Bloom Filler to reduce communication overheads. However, this will inevitably increase false positive rate (fpr) which is a false authentication probability of adversarial messages in trade-off of communication efficiency. While the simple short MAC scheme has the lowest fpr, one cannot choose arbitrary authenticator size. Bloom filter, randomized data structure often used for membership queries, can represent multiple MACs more flexibly with slightly higher fpr. Furthermore, compressed Bloom filter has the same fpr with the short MAC while maintaining its flexibility. Through our detailed analysis, we show that pros and cons of the three schemes are scenario specific. Therefore one can choose appropriate scheme under given parameters to achieve both communication efficiency and security based on our results.

• Communications for Statistical Applications and Methods
• /
• 제16권6호
• /
• pp.1005-1012
• /
• 2009

The area under the curve(AUC) is commonly used as a measure of the receiver operating characteristic(ROC) curve which displays the performance of a set of binary classifiers for all feasible ratios of the costs associated with true positive rate(TPR) and false positive rate(FPR). In the bipartite ranking problem where one has to compare two different observations and decide which one is "better", the AUC measures the quantity that ranking score of a randomly chosen sample in one class is larger than that of a randomly chosen sample in the other class and hence, the function which maximizes an AUC of bipartite ranking problem is different to the function which maximizes (minimizes) accuracy (misclassification error rate) of binary classification problem. In this paper, we develop a way to construct the unified framework for AUC maximizer including support vector machines based on maximizing large margin and logistic regression based on estimating posterior probability. Moreover, we develop an efficient algorithm for the proposed unified framework. Numerical results show that the propose unified framework can treat various methodologies successfully.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제52권1호
• /
• pp.64-75
• /
• 2003

Medical image segmentation is the process by which an original image is partitioned into some homogeneous regions like bones, soft tissues, etc. This study demonstrates an automatic medical image segmentation technique based on independent component analysis. Independent component analysis is a generalization of principal component analysis which encodes the higher-order dependencies in the input in addition to the correlations. It extracts statistically independent components from input data. Use of automatic medical image segmentation technique using independent component analysis under the assumption that medical image consists of some statistically independent parts leads to a method that allows for more accurate segmentation of bones from CT data. The result of automatic segmentation using independent component analysis with square test data was evaluated using probability of error(PE) and ultimate measurement accuracy(UMA) value. It was also compared to a general segmentation method using threshold based on sensitivity(True Positive Rate), specificity(False Positive Rate) and mislabelling rate. The evaluation result was done statistical Paired-t test. Most of the results show that the automatic segmentation using independent component analysis has better result than general segmentation using threshold.

• Imaging Science in Dentistry
• /
• 제50권3호
• /
• pp.209-216
• /
• 2020

Purpose: In India, the age of 14 years is the legal age threshold for child labour. Therefore, in suspected instances of child labour, age assessment plays a crucial role in determining whether a violation of the law on the employment of children has occurred. The aim of this retrospective cross-sectional study was to assess the discriminatory ability of stages of cervical vertebral maturation (CVM) in predicting the legal age threshold of 14 years. Materials and Methods: Routinely taken lateral cephalograms from 408 subjects aged 10 to 18 years were evaluated retrospectively using the CVM stages described by Baccetti et al. Descriptive statistics, accuracy, sensitivity, specificity, positive and negative predictive values, and likelihood ratios were calculated for stages 2, 3, and 4 of CVM. Results: Real age increased as the CVM stage gradually increased. The results of 2×2 contingency tables showed that CVM stage 4 produced an accuracy of 71% and 73%, a false positive rate of 7% and 18%, and a post-test probability of 59% and 68% for boys and girls, respectively. Conclusion: Based on these findings, it can be concluded that the stages of CVM are of limited use for predicting the attainment of the legal age threshold of 14 years. Future studies should investigate whether combinations of skeletal and dental methods could achieve better accuracy and post-test probability.

• 응용통계연구
• /
• 제22권5호
• /
• pp.1085-1096
• /
• 2009

환자-대조군 연관성 연구에서 후보 유전자와 질병이 연관되어 있지 않더라도 인구집단 층화로 인해 가짜 연관성이 발생할 수도 있다. 본 연구에서는 일배체형에 기초한 환자-대조군 연관성 연구에서 인구집단 층화로 인한 가짜 연관성을 해결하기 위한 방법으로, Zaykin 등 (2002)이 제안한 일배체형 경향성 모형에 인구집단 층화에 대한 정보를 추가하고자 한다. Zaykin 등 (2002)의 모형과 제안한 모형에 기초한 일배체형의 유의성 검정에서 인구집단 층화와 인구집단에 대한 관측 오차가 제1종 오류율에 미치는 영향을 모의실험을 통해 살펴보았다. 인구집단이 층화되어 있지만 각 개체가 속한 인구집단을 정확히 알 수 있을 때, Zaykin 등 (2002)의 모형에 기초한 검정은 제1종 오류율을 잘 조절하지 못했지만 본 연구에서 제안한 모형에 기초한 검정은 제1종 오류율을 잘 조절하는 것으로 나타났다. 그러나 인구집단이 층화되어 있고 관측 오차가 존재하면 제안한 모형에 기초한 검정도 제1종 오류율을 조절하지 못하고 명목 유의수준보다 큰 값을 갖는 것으로 나타났다. 따라서 단일염기다형성에 기초한 환자-대조군 연관성 연구와 마찬가지로 일배체형에 기초한 환자-대조군 연관성 연구에서도 인구집단 층화에 대한 정보를 갖고 있다할지라도 그 속에 관측 오차가 존재하면 위양성을 피하기 어렵다는 것을 알 수 있었다.

• Imaging Science in Dentistry
• /
• 제30권3호
• /
• pp.155-158
• /
• 2000

1. 일반방사선사진과 칼라화한 방사선사진의 비교에서 각각 필름에서 진단을 시행할 때 ROC해석법에서는 true positive fraction (TPF), false positive fraction (FPF)를 매개변수로 하고 있으므로 우선 두가지 필름형태에 대해 각각 따로 다음과 같이 평가한다. 2. 판정기준 병변없다 A, 거의 없다 B, 모르겠다 C, 거의 있다 D, 있다 E 먼저 일반방사선사진에서 실제로 병소가 총있는 것이 50, 총없는 것이 50인데 위 판정기준 각각에 대해(equation omitted) 3. 곡선만들기 a.횡축은 FPF 종축은 TPF로 한 그래프를 plot를 한다. sensitivity 17/50 specificity 26/50 accuracy 43/100 b. 곡선만들기 프로그램을 이용하여 곡선을 만들시에는 TPF를 a에 입력하고 PFP를 b에 입력한다. 이 plot을 그릴 수 있는 프로그램은 http://www.members.tripod.co.kr/jdakim 또는 http://www.chosun.ac.kr/∼jdakim의 홈페이지내 공개자료실에서 다운 받으실 수 있습니다. (equation omitted) 이 프로그램에서 입력할 a, b의 값은 (equation omitted) 위와같이 입력하여 얻어진 일반방사선사진에서의 판독 결과 얻어진 곡선이 그래프에서 곡선이 된다. 이와 같은 커브를 컬러화한 사진 판독에서 똑같이 시행하여 ROC곡선(윗곡선)을 만든 다음 두 곡선을 비교하여 아래면적이 더 큰 쪽이 병소 판독에 우수하다고 결론짓는다.