• 한국정보통신학회논문지
• /
• 제6권5호
• /
• pp.684-688
• /
• 2002

침입 탐지 시스템이 가지고 있는 문제점중의 하나는 긍정적 결함(False Positive)이다. 이러한 결함은 침입 탐지 시스템의 성능을 저하시킬 수 있다. 긍정적 결함의 발생원인은 감사데이터 학습단계에서 충분한 학습이 이루어지지 않기 때문에 발생한다. 따라서 본 논문에서는 연관규칙을 탑재한 에이전트에 감사데이터를 학습시키는 방법으로 침입 탐지 시스템을 제안한다.

• 대한방사선기술학회지:방사선기술과학
• /
• 제44권6호
• /
• pp.635-643
• /
• 2021

The results of empirical researches on the diagnosis of lung cancer are insufficient, so it is limited to objectively judge the clinical possibility and utilization according to the accuracy of diagnosis. Thus, this study retrospectively analyzed the lung cancer diagnostic performance of PET-MRI (Positron Emission Tomography-Magnetic Resonance Imaging) by using the decision matrix. This study selected and experimented total 165 patients who received both hematological CEA (Carcinoembryonic Antigen) test and hybrid PET-MRI (¹⁸F-FDG, 5.18 MBq/kg / Body TIM coil. VIVE-Dixon). After setting up the result of CEA (positive:>4 ㎍/ℓ. negative:<2.5㎍/ℓ) as golden data, the lung cancer was found in the image of PET-MRI, and then the SUV_max (positive:>4, negative:<1.5) was measured, and then evaluated the correlation and significance of results of relative diagnostic performance of PET-MRI compared to CEA through the statistical verification (t-test, P>0.05). Through this, the PET-MRI was analyzed as 96.29% of sensitivity, 95.23% of specificity, 3.70% of false negative rate, 4.76% of false positive rate, and 95.75% of accuracy. The false negative rate was 1.06% lower than the false positive rate. The PET-MRI that significant accuracy of diagnosis through high sensitivity and specificity, and low false negative rate and false positive rate of lung cancer, could acquire the fusion image of specialized soft tissue by combining the radio-pharmaceuticals with various sequences, so its clinical value and usefulness are regarded as latently sufficient.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• 제26권6호
• /
• pp.658-665
• /
• 2000

Purpose : The present study was carried out to determine the diagnostic usefulness of bone scan for evaluating jaw bone extension of oral cancer. Materials and Methods : Medical records, preoperative bone scans, computerized tomographic (CT) scans, conventional radiographs, and findings of histopathologic sections of twenty patients who had been treated for oral malignant tumors by a resection of mandible and soft tissue at Chonnam University Hospital from January, 1994 to September, 1999 were analyzed. Results : In 13 cases which showed histopathologically positive, preoperative bone scans were positive in 12 (92.3%) and false negative in 1 (7.7%). Preoperative CT scans were positive in 9 (69.2%) and false negative in 4 (30.8%) of the 13 cases. Preoperative conventional radiographs were positive in 8 (61.5%) and false negative in 5 (38.5%) of the 13 cases. In 7 cases showing negative histopathologic findings, 1 (14.3%) was in CT scans and 2 (28.6%) were false positive in preoperative conventional radiographs. Conclusion : These results suggest that bone scan is more sensitive and reliable method for evaluating jaw bone extension of oral cancer than conventional radiographs or CT scans.

• 한국터널지하공간학회 논문집
• /
• 제26권2호
• /
• pp.129-152
• /
• 2024

터널 내 CCTV를 통한 딥러닝 객체인식 적용에 있어서 터널의 열악한 환경조건, 즉 낮은 조도 및 심한 원근현상으로 인해 오탐지가 대량 발생한다. 이 문제는 객체인식 성능에 기반한 영상유고시스템의 신뢰성 문제로 직결되므로 정탐지 향상과 더불어 오탐지의 저감 방안이 더욱 필요한 상황이다. 이에 본 논문은 딥러닝 객체인식 모델을 기반으로, 오탐지 데이터의 재학습을 통해 오탐지의 저감뿐만 아니라 정탐지 성능 향상도 함께 추구하는 오탐지 학습법을 제안한다. 본 논문의 오탐지 학습법은 객체인식 단계를 기반으로 진행되며, 학습용 데이터셋 초기학습 - 검증용 데이터셋 추론 - 오탐지 데이터 정정 및 데이터셋 구성 - 학습용 데이터셋에 추가 후 재학습으로 이어진다. 본 논문은 이에 대한 성능을 검증하기 위해 실험을 진행하였으며, 우선 선행 실험을 통해 본 실험에 적용할 딥러닝 객체인식 모델의 최적 하이퍼파라미터를 결정하였다. 그리고 본 실험에서는 학습영상 포맷을 결정하기 위한 실험, 반복적인 오탐지 데이터셋의 재학습을 통해 장기적인 성능향상을 확인하기 위한 실험을 순차적으로 진행하였다. 그 결과, 첫 번째 본 실험에서는 추론된 영상 내에서 객체를 제외한 배경을 제거시키는 경우보다 배경을 포함시키는 경우가 객체인식 성능에 유리한 것으로 나타났으며, 두 번째 본 실험에서는 재학습 차수별 독립적으로 오탐지 데이터를 재학습시키는 경우보다 차수마다 발생하는 오탐지 데이터를 누적시켜 재학습 시키는 경우가 지속적인 객체인식 성능 향상 측면에서 유리한 것으로 나타났다. 두 실험을 통해 결정된 방법으로 오탐지 데이터 재학습을 진행한 결과, 차량 객체 클래스는 1차 재학습 이후부터 AP값이 0.95 이상 우수한 추론 성능이 발현되었으며, 5차 재학습까지 초기 추론 대비 약 1.06배 추론성능이 향상되었다. 보행자 객체 클래스는 재학습이 진행됨에 따라 지속적으로 추론 성능이 향상되었으며, 18차 재학습까지 초기 추론대비 2.3배 이상 추론성능이 자가 향상될 수 있음을 보였다.

• 산업공학
• /
• 제21권1호
• /
• pp.33-42
• /
• 2008

To analyze tens of thousands of alarms triggered by the intrusion detections systems (IDS) a day has been very time-consuming, requiring human administrators to stay alert for all time. But most of the alarms triggered by the IDS prove to be the false positives. If alarms could be correctly classified into the false positive and the false negative, then we could alleviate most of the burden of human administrators and manage the IDS far more efficiently. Therefore, we present a new approach based on attribute-oriented induction (AOI) to classify alarms into the false positive and the false negative. The experimental results show the proposed approach performs very well.

• Investigative Magnetic Resonance Imaging
• /
• 제23권1호
• /
• pp.46-54
• /
• 2019

Purpose: The aim of this study was to evaluate the diagnostic performance of a computer-aided detection (CAD) system used with automated breast ultrasonography (ABUS) for suspicious lesions detected on breast MRI, and CAD-false lesions. Materials and Methods: We included a total of 40 patients diagnosed with breast cancer who underwent ABUS (ACUSON S2000) to evaluate multiple suspicious lesions found on MRI. We used CAD ($QVCAD^{TM}$) in all the ABUS examinations. We evaluated the diagnostic accuracy of CAD and analyzed the characteristics of CAD-detected lesions and the factors underlying false-positive and false-negative cases. We also analyzed false-positive lesions with CAD on ABUS. Results: Of a total of 122 suspicious lesions detected on MRI in 40 patients, we excluded 51 daughter nodules near the main breast cancer within the same quadrant and included 71 lesions. We also analyzed 23 false-positive lesions using CAD with ABUS. The sensitivity, specificity, positive predictive value, and negative predictive value of CAD (for 94 lesions) with ABUS were 75.5%, 44.4%, 59.7%, and 62.5%, respectively. CAD facilitated the detection of 81.4% (35/43) of the invasive ductal cancer and 84.9% (28/33) of the invasive ductal cancer that showed a mass (excluding non-mass). CAD also revealed 90.3% (28/31) of the invasive ductal cancers measuring larger than 1 cm (excluding non-mass and those less than 1 cm). The mean sizes of the true-positive versus false-negative mass lesions were $2.08{\pm}0.85cm$ versus $1.6{\pm}1.28cm$ (P < 0.05). False-positive lesions included sclerosing adenosis and usual ductal hyperplasia. In a total of 23 false cases of CAD, the most common (18/23) cause was marginal or subareolar shadowing, followed by three simple cysts, a hematoma, and a skin wart. Conclusion: CAD with ABUS showed promising sensitivity for the detection of invasive ductal cancer showing masses larger than 1 cm on MRI.

• 대한핵의학회지
• /
• 제27권1호
• /
• pp.81-87
• /
• 1993

Retrospective analysis of cholescintigraphy and ultrasonography was done in 76 patients with clinically suspected acute cholecystitis to assess the relative value of the two modalities. Excluding the Patients with obstructive jaundice, the overall results of cholescintigraphy(sensitivity 100%, specificity 95%, false positive rate 5%, false negative rate 0%, accuracy 97%) are nearly identical with or rather superior to those of the ultrasonography(sensitivity 94%, specificity 100%, false positive rate 0%, false negative rate 5%, accuracy 97%). We recommend the cholescintigraphy as the initial modality in patients with clinically suspected acute cholecystitis, and ultrasonography can be used in jaundiced patients to exclude the possibility of the false positive of cholescintigraphy.

• 정보보호학회논문지
• /
• 제24권5호
• /
• pp.809-815
• /
• 2014

최근 정보통신 기술의 발달로 인하여 데이터의 양이 점차 증가하고 있으며, 이에 대한 처리와 관련된 연구가 활발히 진행되고 있다. 주어진 집합 내에 특정 개체의 존재여부를 알기위해 사용되고 있는 블룸필터는 데이터의 공간 활용에 매우 유용한 구조이다. 본 논문에서는 블룸필터에서 발생될 수 있는 오류 확률을 소개한다. 특히 실험실적 분석방법에 의하여 수정된 긍정오류 확률에 대한 일반식을 유도한다. 마지막으로 지금까지 사용되고 있는 블룸필터에 대한 긍정오류확률식과 이에 대한 관련논문을 이용하여 비교, 분석한다.

• 한국멀티미디어학회논문지
• /
• 제20권2호
• /
• pp.144-152
• /
• 2017

Among the technologies for implementing autonomous vehicles, advanced driver assistance system is a key technology to support driver's safe driving. In the technology using the vision sensor having a high utility, various preprocessing methods are used prior to feature extraction for lane detection. However, in the existing methods, the unnecessary lane candidates such as cars, lawns, and road separator in the road area are false positive. In addition, there are cases where the lane candidate itself can not be extracted in the area under the overpass, the lane within the dark shadow, the center lane of yellow, and weak lane. In this paper, we propose an efficient preprocessing method using k-means clustering for image division and the HSV color model. When the proposed preprocessing method is applied, the true positive region is maximally maintained during the lane detection and many false positive regions are removed.

• The Korean Journal of Pain
• /
• 제25권2호
• /
• pp.99-104
• /
• 2012

Background: Previous studies have shown that if performed without radiographic guidance, the loss of resistance (LOR) technique can result in inaccurate needle placement in up to 30% of lumbar epidural blocks. To date, no study has shown the efficacy of measuring the depth of the posterior complex (ligamentum flavum, epidural space, and posterior dura) ultrasonographically to distinguish true and false LOR. Methods: 40 cervical epidural blocks were performed using the LOR technique and confirmed by epidurograms. Transverse ultrasound images of the C6/7 area were taken before each cervical epidural block, and the distances from the skin to the posterior complex, transverse process, and supraspinous ligament were measured on each ultrasound view. The number of LOR attempts was counted, and the depth of each LOR was measured with a standard ruler. Correlation of false and true positive LOR depth with ultrasonographically measured depth was also statistically analyzed. Results: 76.5% of all cases (26 out of 34) showed false positive LOR. Concordance correlation coefficients between the measured distances on ultrasound (skin to ligamentum flavum) and actual needle depth were 0.8285 on true LOR. Depth of the true positive LOR correlated with height and weight, with a mean of $5.64{\pm}1.06cm$, while the mean depth of the false positive LOR was $4.08{\pm}1.00cm$. Conclusions: Ultrasonographic measurement of the ligamentum flavum depth (or posterior complex) preceding cervical epidural block is beneficial in excluding false LOR and increasing success rates of cervical epidural blocks.