• 제목/요약/키워드: ROC with distance

검색결과 23건 처리시간 0.099초

ROC 분석을 이용한 골격성 III급 부정교합의 수평계측방법간 비교연구 (COMPARATIVE STUDY ON THE HORIZOTAL MEASUREMENTS OF SKELETAL CLASS III MALOCCLUSION USING THE ROC ANALYSIS)

  • 최희영;장영일
    • 대한치과교정학회지
    • /
    • 제25권2호
    • /
    • pp.153-163
    • /
    • 1995
  • 본 연구는 III급 부정 교합을 판별하는데 있어, 수평 부조화의 진단에 이용되는 여러 진단 항목들의 진단학적 효율과 타당성을 ROC analysis로 비교하는데 그 목적이 있다. ROC(Receiver Operating Characteristic) analysis는 연속적으로 변하는 cut-off value에서의 sensitivity와 1-specificity에 의해 그려지는 곡선으로서 진단 방법의 타당성을 결정하고, 여러 진단 방법들을 비교하는 분석법으로 알려져 있다. 부정교합자 496명을 대상으로 측모 두부 X-선 계측사진과 진단모형을 이용하여, 진단모형 계측을 통해 부정교합군을 분류하였으며, 이중 III급 부정 교합자는 245명이었다. 측모 두부 X-선계측사진에서 16개의 계측항목을 선정하였으며, 이 계측항목들과 III급 부정교합의 관계를 알아보고자 각도 계측항목에서는 $1^{\circ}$ 간격으로, 선계측항목에서는 1mm의 간격으로 sensitivity와 specificity를 구해 ROC curve를 그렸다. 그리고, 이 계측항목들의 직접적인 비교를 위해 ROC curve 아래의 면적을 계산해냈다. 결과는 다음과 같다. 1. III급 부정교합을 판별하는데 있어, "Wits" appraisal이 다른 계측 항목에 비해 더 나은 진단 효율을 보였다. 2. AB plane angle, ANB angle, App-Bpp distance, AF-BF distance, APDI, N perpendicular to A 와 Pog to N perpendicular의 차이, maxillomandibular differential도 높은 진단 가치를 보였다. 3. 하악골의 위치를 평가하는 계측항목은 중정도의 진단 효율을 보였다. 4. 상악골에 대한 계측항목은 III급 부정교합의 판별에 대한 진단 가치가 낮았다.

  • PDF

Image saliency detection based on geodesic-like and boundary contrast maps

  • Guo, Yingchun;Liu, Yi;Ma, Runxin
    • ETRI Journal
    • /
    • 제41권6호
    • /
    • pp.797-810
    • /
    • 2019
  • Image saliency detection is the basis of perceptual image processing, which is significant to subsequent image processing methods. Most saliency detection methods can detect only a single object with a high-contrast background, but they have no effect on the extraction of a salient object from images with complex low-contrast backgrounds. With the prior knowledge, this paper proposes a method for detecting salient objects by combining the boundary contrast map and the geodesics-like maps. This method can highlight the foreground uniformly and extract the salient objects efficiently in images with low-contrast backgrounds. The classical receiver operating characteristics (ROC) curve, which compares the salient map with the ground truth map, does not reflect the human perception. An ROC curve with distance (distance receiver operating characteristic, DROC) is proposed in this paper, which takes the ROC curve closer to the human subjective perception. Experiments on three benchmark datasets and three low-contrast image datasets, with four evaluation methods including DROC, show that on comparing the eight state-of-the-art approaches, the proposed approach performs well.

NONPARAMETRIC MAXIMUM LIKELIHOOD ESTIMATION OF A CONCAVE RECEIVER OPERATING CHARACTERISTIC CURVE VIA GEOMETRIC PROGRAMMING

  • Lee, Kyeong-Eun;Lim, Johan
    • 대한수학회보
    • /
    • 제48권3호
    • /
    • pp.523-537
    • /
    • 2011
  • A receiver operating characteristic (ROC) curve plots the true positive rate of a classier against its false positive rate, both of which are accuracy measures of the classier. The ROC curve has several interesting geometrical properties, including concavity which is a necessary condition for a classier to be optimal. In this paper, we study the nonparametric maximum likelihood estimator (NPMLE) of a concave ROC curve and its modification to reduce bias. We characterize the NPMLE as a solution to a geometric programming, a special type of a mathematical optimization problem. We find that the NPMLE is close to the convex hull of the empirical ROC curve and, thus, has smaller variance but positive bias at a given false positive rate. To reduce the bias, we propose a modification of the NPMLE which minimizes the $L_1$ distance from the empirical ROC curve. We numerically compare the finite sample performance of three estimators, the empirical ROC curve, the NMPLE, and the modified NPMLE. Finally, we apply the estimators to estimating the optimal ROC curve of the variance-threshold classier to segment a low depth of field image and to finding a diagnostic tool with multiple tests for detection of hemophilia A carrier.

흉부 엑스선 촬영 시 촬영기법에 따른 영상 평가 (Image Evaluation of Projection Method in Chest Radiography)

  • 안병주;이준행
    • 한국방사선학회논문지
    • /
    • 제16권3호
    • /
    • pp.217-223
    • /
    • 2022
  • 본 연구는 엑스선 흉부촬영에서 격자를 이용하여 촬영법과 Air Gap Technique을 이용한 촬영법에서 얻은 영상을 평가하기 위하여 대학병원 영상의학과에서 10년 이상 근무한 방사선사 선생님 5인에게 주관적 평가 ROC와 객관적 평가 SNR를 시행하였다. 분석한 결과는 Cronbach Alpha 값이 0.714로 유의하게 높았다. Air Gap Technique 촬영법에서 팬톰과 상수용체 간격을 20 cm 띄우고 관전압 100 kVp, 관전류와 촬영시간 8 mAs로 설정하고 촬영한 영상을 ROC 평가에서 18점, SNR 평가에서 6.149로 가장 높은 점수를 얻었다. 또한 격자를 이용하여 촬영법에서는 팬톰과 상수용체 간격을 15 cm 띄우고 관전압 110 kVp, 관전류와 촬영시간을 8 mAs에서 촬영 시 ROC 평가점수가 19점과 객관적평가인 신호대잡음비(SNR)가 6,622으로 가장 높은 점수를 받았다. 이에 격자를 이용하여 나타나는 격자의 cut-off 현상으로 인하여 재촬영뿐만 아니라 판독지연, 환자의 피폭선량 증가, 기계적 수명단축 등의 단점을 벗어나는 Air Gap Technique 촬영법을 활용한다면, 중환자실 및 응급환자를 위한 이동형 엑스선 촬영장비를 사용할 경우를 포함하여 흉부 촬영 시에 많은 도움이 되리라 사료된다.

A Comparison of Urban Growth Probability Maps using Frequency Ratio and Logistic Regression Methods

  • Park, So-Young;Jin, Cheung-Kil;Kim, Shin-Yup;Jo, Gyung-Cheol;Choi, Chul-Uong
    • 한국조경학회지
    • /
    • 제38권5_2호
    • /
    • pp.194-205
    • /
    • 2010
  • To predict urban growth according to changes in landcover, probability factors werecal culated and mapped. Topographic, geographic and social and political factors were used as prediction variables for constructing probability maps of urban growth. Urban growth-related factors included elevation, slope, aspect, distance from road,road ratio, distance from the main city, land cover, environmental rating and legislative rating. Accounting for these factors, probability maps of urban growth were constr uctedusing frequency ratio (FR) and logistic regression (LR) methods and the effectiveness of the results was verified by the relative operating characteristic (ROC). ROC values of the urban growth probability index (UGPI) maps by the FR and LR models were 0.937 and 0.940, respectively. The LR map had a slightly higher ROC value than the FR map, but the numerical difference was slight, with both models showing similar results. The FR model is the simplest tool for probability analysis of urban growth, providing a faster and easier calculation process than other available tools. Additionally, the results can be easily interpreted. In contrast, for the LR model, only a limited amount of input data can be processed by the statistical program and a separate conversion process for input and output data is necessary. In conclusion, although the FR model is the simplest way to analyze the probability of urban growth, the LR model is more appropriate because it allows for quantitative analysis.

Statistical Fingerprint Recognition Matching Method with an Optimal Threshold and Confidence Interval

  • Hong, C.S.;Kim, C.H.
    • 응용통계연구
    • /
    • 제25권6호
    • /
    • pp.1027-1036
    • /
    • 2012
  • Among various biometrics recognition systems, statistical fingerprint recognition matching methods are considered using minutiae on fingerprints. We define similarity distance measures based on the coordinate and angle of the minutiae, and suggest a fingerprint recognition model following statistical distributions. We could obtain confidence intervals of similarity distance for the same and different persons, and optimal thresholds to minimize two kinds of error rates for distance distributions. It is found that the two confidence intervals of the same and different persons are not overlapped and that the optimal threshold locates between two confidence intervals. Hence an alternative statistical matching method can be suggested by using nonoverlapped confidence intervals and optimal thresholds obtained from the distributions of similarity distances.

Utility of False Profile View for Screening of Ischiofemoral Impingement

  • Kwak, Dae-Kyung;Yang, Ick-Hwan;Kim, Sungjun;Lee, Sang-Chul;Park, Kwan-Kyu;Lee, Woo-Suk
    • Hip & pelvis
    • /
    • 제30권4호
    • /
    • pp.219-225
    • /
    • 2018
  • Purpose: Ischiofemoral impingement (IFI)-primarily diagnosed by magnetic resonance imaging (MRI)-is an easily overlooked disease due to its low incidence. The purpose of this study was to evaluate the usefulness of false profile view as a screening test for IFI. Materials and Methods: Fifty-eight patients diagnosed with IFI between June 2013 and July 2017 were enrolled in this retrospective study. A control group (n=58) with matching propensity scores (age, gender, and body mass index) were also included. Ischiofemoral space (IFS) was measured as the shortest distance between the lateral cortex of the ischium and the medial cortex of lesser trochanter in weight bearing hip anteroposterior (AP) view and false profile view. MRI was used to measure IFS and quadratus femoris space (QFS). The receiver operating characteristics (ROC), area under the ROC curve (AUC) and cutoff point of the IFS were measured by false profile images, and the correlation between the IFS and QFS was analyzed using the MRI scans. Results: In the false profile view and hip AP view, patients with IFI had significantly decreased IFS (P<0.01). In the false profile view, ROC AUC (0.967) was higher than in the hip AP view (0.841). Cutoff value for differential diagnosis of IFI in the false profile view was 10.3 mm (sensitivity, 88.2%; specificity, 88.4%). IFS correlated with IFS (r=0.744) QFS (0.740) in MRI and IFS (0.621) in hip AP view (P<0.01). Conclusion: IFS on false profile view can be used as a screening tool for potential IFI.

심혈관계질환의 발병 위험을 선별할 수 있는 적정 허리둘레의 임계점 (Optimal Waist Circumference for the Prevention of Cardiovascular Disease)

  • 백인경;신철
    • 대한지역사회영양학회지
    • /
    • 제15권2호
    • /
    • pp.275-283
    • /
    • 2010
  • There are few studies reporting optimal waist circumference that can be utilized to prevent the incidence of cardiovascular disease (CVD). We evaluated the association of waist circumference and waist and hip circumference ratio (WHR) with incident cases of CVD developed over 6 years in a population-based prospective study including Korean adults. Analyses for receiver-operating characteristic (ROC) curve were performed with data for 1,733 men and 1,579 women who were aged 40 to 69 years and were free of a physician-diagnosis of CVD at baseline. Information on the diagnosis of CVD was periodically reported using interviewer-administered questionnaires and anthropometric measures were obtained by biennial health examinations. We newly identified 77 cases of CVD during a follow-up period between 2003 and 2008. On the basis of measures of diagnostic accuracy including minimum distance to ROC curve and Youden index, waist circumference of 85 cm for men, in particular for male nonsmokers, and of 80 cm for women and WHR of 0.88 to 0.90 for men and of 0.83 for women were found to be optimal cutoff points to identify individuals at CVD risks. The study also found that the use of the suggested optimal values for waist circumference show higher sensitivity and lower specificity compared with 90 cm for men and 85 cm for women, which are waist cutoff points given by the Korean Society for the Study of Obesity to define abdominal obesity for Korean adults. Although lower cutoff points of waist circumference (83 cm) and WHR (0.87) were observed to be optimal for male smokers compared with male nonsmokers, whether suggesting waist cutoff points specific to smokers is needed warrants further studies. After taking into account other cardiovascular risk factors including smoking, men with waist circumference of 85 cm or greater and women with 80 cm or greater were at an increased risk of CVD. Thus, these cutoff points of waist circumference may be able to capture more individuals at CVD risks contributing to the prevention of future development of CVD.

거리 기반 유사도 측정을 통한 유방 초음파 영상의 내용 기반 검색 컴퓨터 보조 진단 시스템에 관한 연구 (A Study of CBIR(Content-based Image Retrieval) Computer-aided Diagnosis System of Breast Ultrasound Images using Similarity Measures of Distance)

  • 김민정;조현종
    • 전기학회논문지
    • /
    • 제66권8호
    • /
    • pp.1272-1277
    • /
    • 2017
  • To assist radiologists for the characterization of breast masses, Computer-aided Diagnosis(CADx) system has been studied. The CADx system can improve the diagnostic accuracy of radiologists by providing objective information about breast masses. Morphological and texture features were extracted from the breast ultrasound images. Based on extracted features, the CADx system retrieves masses that are similar to a query mass from a reference library using a k-nearest neighbor (k-NN) approach. Eight similarity measures of distance, Euclidean, Chebyshev(Minkowski family), Canberra, Lorentzian($F_2$ family), Wave Hedges, Motyka(Intersection family), and Cosine, Dice(Inner Product family) are evaluated by ROC(Receiver Operating Characteristic) analysis. The Inner Product family measure used with the k-NN classifier provided slightly higher performance for classification of malignant and benign masses than those with the Minkowski, $F_2$, and Intersection family measures.

Dimensional Stability of Bentwoods by Treatment Conditions

  • Jung, In-Suk;Lee, Weon-Hee;Chang, Jun-Pok;Bae, Hyun-Mi
    • Journal of the Korean Wood Science and Technology
    • /
    • 제30권3호
    • /
    • pp.85-90
    • /
    • 2002
  • This study was carried out to investigate the dimensional stability of bentwoods by three treatments: steaming, urethane varnish coating, and polyethylene glycol (PEG) treatment. Bentwood processing employed a bending-jig with only 4 cm radius of curvature (ROC). The used species were bitter wood (Picrasma quassioides), painted maple (Acer mono), and birch (Betula schmidtiii). The bending properties of these are well-known in bentwood production (Jung et al., 2002). The bentwoods were treated repeated at room temperature [20℃, RH 80% (12 hours) and 40℃ under RH 10% (12 hours)]. To estimate the dimensional stability of bentwoods, we measured the radius of curvature and end-distance. The best results could be attained with PEG treatment. Steaming was the worst treatment. Comparing the properties of the different species, the dimensional stability of bitter wood was excellent. It was concluded that the steaming treatment was unsuitable for dimensional stability of bentwoods.