Showing papers on "False positive paradox published in 1997"

Distressed or relieved? Psychological side effects of breast cancer screening in the Netherlands

Abstract: STUDY OBJECTIVES: To assess the psychological impact of mammographic screening on women with non-malignant outcomes after attending the Netherlands' National Breast Cancer Screening Programme. DESIGN: During one year all women with false positive test results (95) in a screening area were invited for the study. Each false positive was matched with two women with normal mammograms with respect to age and municipality. A random reference group of 400 was drawn from the female population in an area not yet included in the screening programme. Experiences with screening and psychological status of subjects were assessed 8-10 weeks after screening (T1) and again after six months (T2), by interviews as well as questionnaires. References completed two questionnaires with a six months' interval. PARTICIPANTS: 74 (78%) women with false positive outcomes and 113 (59%) women with negative outcomes participated at T1, of these 65 (88%) and 105 (93%) at T2, respectively; 238 references returned questionnaires at T1 (59%), of these 143 (60%) at T2. MAIN RESULTS: At 8-10 weeks after the screening, the women who received false positive test results scored higher on most of the variables indicating psychological disfunctioning than women with normal mammograms, but did not notably differ on the same variables from the non-screened reference group. Women with normal mammograms had the lowest scores on all the variables in the study at both assessments. The same situation was observed six months later. Although 61% of the women who received false positive mammograms reported that they had experienced the "false alarm" as a stressful event, this experience had apparently no adverse effects on their psychological functioning, as assessed 8-10 weeks after screening. CONCLUSIONS: Overall, breast screening is not likely to generate adverse psychological effects in "healthy" women, even if the outcome is false positive. Differences in psychological functioning between false positives and negatives are more likely ascribable to feelings of relief in the negative group than to raised anxiety and distress in the false positive group.

The psychosocial consequences of mammography.

Abstract: Increasing numbers of mammograms being performed in the United States will be accompanied inevitably by an increasing number of false positives. According to reliable estimates from a survey of radiology facilities, U.S. women in their forties experience close to one million false positive mammograms every year. To determine the impact of false positive mammograms and the broader psychological impact of mammography, we conducted literature searches of Medline, CancerLit, and PsycInfo. We identified nine studies examining the impact of false positive mammograms. Most found short-term increases in such psychological measures as anxiety, distress, and intrusive thoughts. One study found substantial effects on these measures three months after an abnormal mammogram. Another study found an 18-month impact on anxiety. Few studies have used behavioral outcomes, but one reported overpractice of breast self-exam among women who had received false positive results. Another found no reduction in adherence to mammography among women who have had an abnormal test. The more general mammography literature suggests that many women are anxious about mammography before the exam; women with lower levels of education, African Americans, and women with a family history of breast cancer may be more vulnerable to distress. Unfortunately, this literature suffers major limitations, such as small sample sizes, inconsistent and sometimes inappropriate measures, variations in the time frames for measurement, few studies with women aged 40-49, and a paucity of U.S. research. More research is needed to characterize at-risk women and to test interventions designed to reduce the negative impact of abnormal mammograms. Improved communication is also needed throughout the entire mammography process.

Assaying system for illicit substances using intentional false positives to initially preserve anonymity

Abstract: An assaying system for determining the presence of particular illicit substances in human physiological fluid, consisting of a collection cup, a collection cup lid, and a test card with assaying means, quality control means and adulteration detection means which is secured to the collection cup lid. The assaying means contemplate the addition of control positives such that a pre-determined number of assaying systems would be configured so that a false positive indication would be given by the assaying means in addition to any true positive results which may be present. Accordingly, it would initially be impossible to distinguish true test-positives from control positives, and thus the anonymity of a true test-positive donor individual would be protected. All donor individuals testing negative would be immediately discernible.

The correction for bias in prevalence estimation with screening tests.

Abstract: The concern that the South Oaks Gambling Screen (SOGS) and other screening tests have a relatively high rate of false positive errors which results in overestimation of the true prevalence in general population studies is shown to be unfounded. False positives are seen to be a necessary but not sufficient condition for overestimation. It is demonstrated that the proper research question is whether the sample prevalence estimator is biased, and, if so, in which direction. One solution to the problem of bias is shown to depend on the availability of estimates of the error rates of the test.

Trade-off between false positives and false negatives in the linkage analysis of complex traits.

Abstract: This study examines the issue of false positives in genomic scans for detecting complex trait loci using subpair linkage methods and investigates the trade-off between the rate of false positives and the rate of false negatives. It highlights the tremendous cost in terms of power brought about by an excessive control of type I error and, at the same time, confirms that a larger number of false positives can occur otherwise in the course of a genomic scan. Finally, it compares the power and rate of false positives obtained in preplanned replicated studies conducted using a liberal significance level to those for single-step studies that use the same total sample size but stricter levels of significance. For the models considered here, replicate studies were found more attractive as long as one is willing to accept a trade-off, exchanging a much lower rate of false negatives for a slight increase in the rate of false positives.

Validating the defect detection performance advantage of group designs for software reviews: report of a laboratory experiment using program code

Abstract: It is widely accepted that software development technical reviews (SDTRs) are a useful technique for finding defects in software products. Recent debates centre around the need for review meetings (Porter and Votta 1994, Porter et al 1995, McCarthy et al 1996, Lanubile and Visaggio 1996). This paper presents the findings of an experiment that was conducted to investigate the performance advantage of interacting groups over average individuals and artificial (nominal) groups. We found that interacting groups outperform the average individuals and nominal groups. The source of performance advantage of interacting groups is not in finding defects, but rather in discriminating between true defects and false positives. The practical implication for this research is that nominal groups constitute an alternative review design in situations where individuals discover a low level of false positives.

Influence of postmortem changes on immunohistochemical reactions in skin.

Abstract: The influence of postmortem damage of tissues on the immunohistochemical diagnosis of wound age has not as yet been clarified. We utilized antibodies against the proteinase inhibitors α-1-antichymotrypsin and α-2-macroglobulin, fibronectin and lysozyme to study samples of skin which had been intact intravitally, but were damaged postmortem either by autolysis or compression with a surgical clamp at the time of dissection. Even in the absence of autolysis, antibodies against the proteinase inhibitors and fibronectin exhibited staining of tissue margins. Autolysis caused an increase in false positive results. In contrast, antibodies against lysozyme did not give false positive staining. There were no antigens sensitive to postmortem clamping and false positive results were not observed. Antibodies against proteinase inhibitors are not useful for the diagnosis of wound age because of a high number of false positive reactions in marginal areas. Fibronectin also showed false positive band-shaped staining patterns at the tissue margin. In addition, autolytic processes increase the number of false positives. The antibody against lysozyme is much less sensitive to autolysis and no false positive reactions were observed in our series of tests.

Contribution of 99mTc-MIBI scintimammography to the diagnosis of non-palpable breast lesions in relation to mammographic probability of malignancy.

Abstract: The low positive predictive value of mammography results in unnecessary biopsies. We present a prospective evaluation on the contribution of 99m Tc-MIBI Scintimammography (SMM) to the diagnosis of breast cancer in 41 patients with non-palpable breast lesions detected by mammography. In all cases mammographical findings were indicative of biopsy and according to the probability of malignancy they were classified into three groups: high probability (17), intermediate (15), and low (9). There were 22 malignant lesions and 19 benign. In the high probability group, MIBI-SMM changed the only false positive into true negative, and showed 2 false negatives. In the intermediate group, MIBI-SMM changed 7 of 11 false positives on mammography into true negatives, and showed 1 false negative. In the low probability group MIBISMM changed 3 of the 7 false positive into true negatives without false negatives. In the 24 patients included in the intermediate and low probability groups, 10 of the 18 false positives were changed into true negatives by MIBI-SMM at the expense of I false negative. The addition of SMM may to reduce up to 55% of the number of unnecesary biopsies in non-palpable breast lesions.

Investigation of regularized neural networks for the computerized detection of mass lesions in digital mammograms

Abstract: Computerized schemes are currently being developed at the University of Chicago to detect mass lesions in digital mammograms. Artificial neural networks play an important role in the detection of masses. Currently, features are extracted from potential lesion areas and sent through a neural network to decide whether the area is to be called a true lesion or a false detection. One of the most difficult aspects of dealing with artificial neural networks is to train them without over-training; in other words, to take both the bias and variance into account when training. Typically, an early stopping technique is employed; that is, the neural network is tested on an independent data set and training is stopped when the the performance on this independent data set is maximized. In this paper the effectiveness of regularization is evaluated as a technique to minimize over-training. Regularization adds an extra term to the cost-function used in neural network training that penalizes over-complex results. The results of simulation studies will be presented along with results obtained using data of actual lesions and false positives from the authors' computerized mass detection scheme.

The effect of Gaussian error in object recognition

Abstract: In model based recognition, the goal is to locate an instance of one or more known objects in an image. The problem is compounded in real images by the presence of clutter, occlusion, and sensor error, which can lead to "false negatives", failures to recognize the presence of the object, and "false positives", in which the algorithm incorrectly identifies an occurrence of the object. The probability of either event is affected by parameters within the recognition algorithm, which are almost always chosen in an ad-hoc fashion. The effect of the parameter values on the likelihood that the recognition algorithm will make a mistake are usually not understood explicitly. To address the problem, we explicitly model the noise that occurs in the image. In a typical recognition algorithm, hypotheses about the position of the object are tested against the evidence in the image, and an overall score is assigned to each hypothesis. We use a statistical model to determine what score a correct or incorrect hypothesis is likely to have, and use standard binary hypothesis testing techniques to distinguish correct from incorrect hypotheses. Using this approach, we can compare algorithms and noise models, and automatically choose values for internal system thresholds to minimize the probability of making a mistake.

Validating the defect detection performance advantage of group designs for software reviews: report of a replicated experiment

Abstract: It is widely accepted that software development technical reviews (SDTRs) are a useful technique for finding defects in software products. The normative SDTR literature assumes that group reviews are better than individual reviews. Recent debates centre around the need for review meetings. This paper presents the findings of a replicated experiment that was conducted to investigate whether group review meetings are needed and why. We found that an interacting group is the preferred choice over the average individual and artificial (nominal) groups. The source of performance advantage of interacting groups is not synergy as was previously thought, but rather in discriminating between true defects and false positives identified by individual reviewers. As a practical implication, nominal groups may be an alternative review design in situations where individuals exhibit a low level of false positives.

The IPCS Collaborative Study on Neurobehavioral Screening Methods: VII. Summary and conclusions

Abstract: In the International Programme on Chemical Safety (IPCS) Collaborative Study on Neurobehavioral Screening Methods, eight participating laboratories used a standard battery of behavioral tests to determine, in rats, the effects of seven representative chemicals following acute and repeated dosing. The results of the collaborative study indicate good agreement across laboratories with regard to the data collected in vehicle controls. It was clear, however, that some behavioral measures had significantly more variability than other tests. The laboratories also demonstrated the ability to detect known neurotoxic chemicals and identify profiles of effects that differed from non-neurotoxic agents. The results of the study suggest that appropriate training of personnel is crucial to ensure the reliability of the test battery. The results also underscore the importance of dose selection in behavioral screening studies, since it is sometimes difficult to determine the specificity of behavioral changes in animals receiving high doses of some chemicals. The collaborative study also emphasizes the need to utilize a battery of tests in screening a wide range of potential neurotoxic agents. Analysis of data from such studies poses unique challenges due to the large number of tests and test times, and the consequent possibility of false positives. Some statistical concerns may be alleviated by grouping the results from tests that measure similar functions into neurobiological domains. Although this approach improves confidence in the biological relevance of chemical-induced changes in behavior, it may also lead to false negatives. The exploration of other statistical approaches to analyze data from experiments using a test battery is encouraged. Nevertheless, results of the collaborative study strongly support the use of behavioral tests in hazard identification.

Wavelet snake for classification of nodules and false positives in digital chest radiographs

Abstract: The authors have developed a technique using a wavelet-based snake model for classification of nodules and false positives reported by their computer-aided diagnosis (CAD) scheme for detection of pulmonary nodules in digital chest radiographs. In this technique, the boundary of a nodule is first approximated by the multiscale edge representation. These multiscale edges are used to "guide" the snake to estimate the true boundary of the nodule. The deformation of the snake is performed by changing the wavelet coefficients using a gradient descent algorithm. The degree of overlap between the fitted snake and the multiscale edges was calculated as a measure for classification of nodules and false detections. A set of ROIs consisting of 90 nodules and 152 false positives were used for evaluation of the authors' method by means of the receiver operating characteristic (ROC) analysis. Their method yielded an area under the ROC curve of 0.74 in classification performance, which reduced 15% of the false positives with sacrifice of only one module.

Adaptive feature analysis of false positives for computerized detection of lung nodules in digital chest images

Abstract: To assist radiologists in diagnosing early lung cancer, we have developed a computer-aided diagnosis (CAD) scheme for automated detection of lung nodules in digital chest images. The database used for this study consisted of two hundred PA chest radiographs, including 100 normals and 100 abnormals. Our CAD scheme has four basic steps, namely, (1) preprocessing, (2) identification of initial nodule candidates (rule-based test #1), (3) grouping of initial nodule candidates into six groups, and (4) elimination of false positives (rule-based test #2 - #5 and artificial neural network). Our CAD scheme achieves, on average, a sensitivity of 70%, with 1.7 false positives per chest image. We believe that this CAD scheme with its current performance is ready for clinical evaluation.© (1997) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

False negative rate. A misnomer, misunderstood and misused.

Abstract: By rigorous statistical standards, the expression false negative rate is a misnomer because the concept of "rate" is associated with the rapidity of change of phenomena per unit of time or other variable (e.g., temperature). Uncertainty about the term false negative rate resulted in the introduction of similar but equally dubious expressions, including false negative fraction and miss rate. The statistically correct term in the practice of cytopathology is false negative proportion because we have here a fraction in which the numerator (number of false negative cases) is a part of the denominator (number of true positive cases plus number of false negative cases). This is the classic definition of the term proportion.

Radiologists' ability to discriminate computer-detected true and false positives from an automated scheme for the detection of clustered microcalcifications on digital mammograms

Abstract: There is evidence that computer-aided diagnosis (CAD) can be used to improve radiologists' performance. However, one of the potential drawbacks of CAD is that a computer-detected false positive may induce a false positive by a radiologist. To examine this issue, we performed two experiments to compare radiologists' false positives with those of the computer and to determine radiologists' ability to discriminate between the computer's true- and false-positive detections. In the first experiment, radiologists were shown 50 mammograms and on each film were asked to indicate 3 regions that could contain clustered microcalcifications, and using a 100-point scale, to give their level of confidence that microcalcifications were present in the region. In the second experiment, the radiologists were shown regions-of-interest, printed on film, containing either a computer-detected true cluster or a computer- detected false positive. The radiologists gave their confidence that there were actual clustered microcalcifications present. There was less than 1% overlap between false positives by the computer and radiologists. Furthermore, based on ROC analysis, radiologists were able to discriminate between computer true and false positives.© (1997) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Avoiding cultural false positives

Abstract: business \" Frontiers in Electronic Media \" figure 21. See page 61 I International studies have the potential to introduce a particular bias that can falsely attribute differences in user needs to geography or culture. We have coined the term \" cultural false positives \" to describe study results that mistakenly ascribe differences in users' goals and tasks to culture—differences which are actually variations in user type or segmentation. cial experiences and expectations that create a need for unique or highly customized user interfaces for computer products. [2] We question this premise for the development of computer products. Clearly, some differences arise due to country and region, such as language, legal regulations, paper formats, and responses to marketing campaigns. We call these differences \" locale differences, \" and recognize that a computer system must support these through classic internationalization to ensure a minimum level of usability. According to Kano [5], internationalization—also called globalization—is \" the process of developing a program core whose feature design and code design does not make assumptions based on a single language or locale and whose source code base simplifies the creation of different language editions of a program. \" Are the non-locale differences we see in user needs—such as size and color—funda-mental differences based on a user's geogra-phy? Our international studies have revealed that differences beyond locale are often due to variations in user type and segmentation, and that what are categorized as cultural differences do not impact the human-computer interaction model. Based on our involvement in a variety of product development teams, we believe the differences can most often be attributed to user type, tasks, or goals. The term also serves as a reminder of the potential bias that can result from these studies. To avoid cultural false posi-tives, project teams need to iteratively evaluate how their bias is influencing their current design—if at all. A team can avoid cultural false positives by maintaining its focus on the basics of design, including users and their tasks, goals, and environments.

CAD system for full-digital mammography and its evaluation

Abstract: The purpose of this study is to develop a clinical intelligent workstation for computer-aided diagnosis (CAD) of breastcancer using full digital mammography. It consists of a clinical workstation and Fuji Computed Radiography 9000 System. Newimage processing methods to extract tumor masses and clustered microcalcifications have been developed and implemented inthe CAD system. A new filter called Iris Filter has been developed to detect tumor candidates. It realizes reliable detection oftumor candidates regardless of their sizes and their contrast against their background on mammograms. And a new methodbased on mathematical morphology has been developed to detect microcalcifications. It is adaptive to the imaging conditions ofmammograms. One thousand, two hundred and twelve CR images, which include 240 malignant tumors, were used to test theperformance of the system. The sensitivity for malignant tumors was 90. 5% and the average number of false positives per imagewere only 1.3. The true positive detection rate for clustered microcalcifications was 89.2% and the average number of falsepositives per image were 0.36. The high true positive rates and the low false positive detection characterize the proposed full-digital CAD system, which shows the possibility of practical application of computer aided diagnosis of breast cancer.Keywords: breast cancer, mammography, CAD, malignant tumors, microcalcijIcations, irisfilter. morphological processing

Detection and classification of mines via discriminant features and borrowed strength

Abstract: In this paper we introduce a method for detecting landmines from remotely sensed data. Our primary goal is to minimize the number of false negatives, while keeping the number of false positive very low.We concentrate on two different phases of the processing: detection and classification. We construct a feature based detector which utilizes some basic signal processing techniques to produce a detection map which has very few false negatives, and quite a few false positives. The detection phase of this paper concentrates on a background equalized matched filter, along with some measures of symmetry. After the detection phase we will construct a classifier for the problem which concentrates on estimating the density distributions of the pixel values. We will show that a variation of 'Borrowed Strength' as introduced by Priebe, significantly improves the classification process. This classification technique will reduce the number of false positives, without significantly increasing the number of false negatives. The procedure is demonstrated on passive multispectral mine data generously provided by the NSWC Dahlgren Division. Coastal Systems Station, Panama City, Florida. In conclusion, we propose a method for improving the performance of this type of classification. We believe that the proposed method shows a great deal of promise, and will be of use in many situations.

Atypicality Revisited: Further Data on the Effectiveness of the Mahalanobis Distance in Down's Syndrome Screening

Abstract: Further data suggesting that assessment of atypicality may be a valuable addition to Down's syndrome screening is presented, based on results from 1521 unaffected pregnancies and 190 chromosomally abnormal pregnancies. For a 1 % increase in false positive rates (over the Down's screening-related false positives), it is estimated that the detection of non-trisomy 21 abnormalities may be enhanced by approximately 40%. These false positive and detection rates are similar to previous reports for specific trisomy 18 screens. Further work is necessary to prove whether a specific trisomy 18 screen, or an atypicality screen is preferable.

Mammographie und Mammasonographie: In welcher Reihenfolge sollten die Untersuchungen erfolgen?

Abstract: PURPOSE To compare primary mammography diagnosis (ultrasound report available) with primary ultrasound diagnosis (mammography report available). METHODS 89 preoperative patients with suspicious lesions were included. Mammography and ultrasound of all patients were evaluated by two independent experienced readers under clinical conditions. The reports of the complementary modality were available to both observers. Lesion evaluation was done on a per breast basis, in cases of multiple lesions in respect of the lesion with the greatest risk of malignancy. RESULTS 39 benign and 59 malignant lesions were found. Primary mammography and primary ultrasound yielded 3 and 8 false positives and 10 and 13 false negatives. Concerning the palpable lesions (n = 59), primary mammography and primary ultrasound had no and 4 false positives and 7 and 8 false negatives, respectively, for the non palpable lesions, the figures were 3 and 4 false positive and 3 and 5 false negatives. CONCLUSIONS Mammography remains the method of first choice in early detection of breast cancer, whereas breast ultrasound should be performed after and in knowledge of the mammogram, in consideration of the known indications (equivocal palpable lesion and mammographic opacity, dense breast).

Evaluation of atrazine positive and false positive immunoassay detections in ground water

Abstract: False positive responses on an atrazine (6‐chloro‐N‐ethyl‐N'‐(l‐methylethyl)‐1, 3, 5‐triazine‐2, 4‐diamine) immunoassay kit were investigated to explain possible causes for these occurences. Ground water samples were evaluated with the immunoassay kit and positive responses (>0.20 μg L‐1) were confirmed using gas chromatography/mass spectrometry (GC/MS). Non‐confirming samples (false positives) were analyzed for seven additional compounds on GC. Resulting GC/MS and GC analyses showed that 70% of the false positives could be attributed to two compounds. Prometon (6‐methoxy‐N,N'‐bis(l‐methylethyl)‐1, 3, 5‐triazine‐2, 4‐diamine) was responsible for the majority (64%) of the false positive responses The atrazine metabolite, deethylatrazine (2‐chloro‐4‐amino‐6‐isopropylamino‐1, 3, 5‐triazine), was responsible for the other 6% of the false positives measured. Unattributed false positives (30%) were probably due to an overestimation of pesticide concentrations in the kit's lower detection range.