Screening mammography in community practice: Positive predictive value of abnormal findings and yield of follow-up procedures

Applied Research Branch, National Cancer Institute, Bethesda, MD 20892-7344, USA.
American Journal of Roentgenology (Impact Factor: 2.73). 01/1996; 165(6):1373-7. DOI: 10.2214/ajr.165.6.7484568
Source: PubMed


The purpose of this study was to gather from 50 community mammography practices that were included in the National Survey of Mammography Facilities data concerning abnormal findings on screening mammograms to determine the frequency of various recommendations made for patients who had abnormal findings and to compare these recommendations with the frequency with which the procedures were actually performed. We also determined the positive predictive value of screening mammograms (the number of cancers detected per 100 abnormal findings) and the yield (the number of cancers detected per 100 procedures done) of various diagnostic procedures done because of abnormal findings.
We identified 1717 screening mammograms done in the last half of 1991 that had abnormal findings. Radiologic recommendations and follow-up procedures, including repeat standard (screening) mammography, additional mammographic views, sonography, clinical breast examination, needle aspiration, needle biopsy, and open biopsy, were identified for all of the cases from the radiologic records, and follow-up data were obtained from referring physicians. The positive predictive value and yield in the National Survey of Mammography Facilities were compared with data from the mammography screening practice of the University of California at San Francisco (UCSF), a facility noted for its clinical efficiency.
We estimate that 11% of all screening mammograms resulted in a recommendation for further diagnostic procedures. These 1717 mammograms with abnormal findings led to the following recommendations and procedures: repeat standard (screening) mammography, 610 (recommended)/635 (performed); additional mammographic views, 785/707; sonography, 400/345; biopsy, 189/229; and needle aspiration, 21/51. More procedures were done than were recommended in some cases because the results of certain procedures often led to the performance of other, additional procedures. The positive predictive value for screening examinations with abnormal findings was 3.5%, and the yield for open biopsy was 21%. In the UCSF data base, the positive predictive value for examinations with abnormal findings was 10%, and the yield for open biopsy was 34%.
The positive predictive value for examinations with abnormal findings and the yield for diagnostic procedures performed as a result of abnormal findings in 50 community radiologic facilities were higher than those reported in some earlier studies, a fact that raised concern about the induced cost of screening mammography. However, these values were low compared with those in the UCSF data base. This fact was particularly true of repeat standard (screening) mammography.

Download full-text


Available from: Larry Kessler,
  • Source
    • "X-ray mammography is currently the gold standard medical imaging tool for breast tumour detection [4]. However, it has a number of limitations, such as missing approximately 10% -30% of breast cancers [4], discomfort and unpleasant pain to the patient during operation [5], it is can be unsuitable for young women and breast feeding women [6] and the ionizing properties of X-rays restrict the frequency of screening [7]. These limitations of X-Ray mammography motivate the development of new alternate imaging modalities. "

    Journal of Biomedical Science and Engineering 08/2013; Volume 6(No.8):823-833. DOI:10.4236/jbise.2013.68100
  • Source
    • "However, B-mode images are not always reliable because benign and malignant lesions could have similar B-mode appearances. As a result, currently, almost 80% of the biopsies carried out after radiological referral turn out to be benign [2], [5], [6]. Thus, there exists a need for reducing the number of breast biopsies and improving ultrasound-based diagnosis. "
    [Show abstract] [Hide abstract]
    ABSTRACT: In this work, we report the use of ultrasound RF time series analysis for separating benign and malignant breast lesions with similar B-mode appearance. The RF time series method is versatile and requires only a few seconds of imaging. We have employed the spectral and fractal features of ultrasound RF time series and used support vector machines with leave-one-patient-out cross validation of the classification. We have also produced cancer probability maps, by estimating the posterior malignancy probability of regions of size 1 mm 2 in the suspicious lesions. The first 12 patient cases of our ongoing study are reported here. Pathologic analysis of the cores using ultrasound guided needle biopsy confirmed the tissue type. We report an area under receiver operating characteristic curve of 0.82. We were able to successfully classify 6 out of 7 patients with malignant breast lesions and 4 out of 5 patients with benign lesions, with success defined as correct classification of at least 75% of the 1 mm 2 regions in the area of the lesion. The above findings suggest that ultrasound time series along with support vector machines can help in differentiating malignant from benign breast lesions.
  • Source
    • "It is reported that in the US, only about 20% of women who have biopsies turn out to have cancer [9]. While only about 3.5% of abnormal screening mammograms interpreted by community radiologists reveal cancer, subspecialty radiologists have a significantly higher positive predictive value (PPV) [8]. Community radiologists also have a lower sensitivity resulting in missed breast cancers. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Screening mammography is the most effective means for early detection of breast cancer. Although general rules for discriminating malignant and benign lesions exist, radiologists are unable to perfectly detect and classify all lesions as malignant and benign, for many reasons which include, but are not limited to, overlap of features that distinguish malignancy, difficulty in estimating disease risk, and variability in recommended management. When predictive variables are numerous and interact, ad hoc decision making strategies based on experience and memory may lead to systematic errors and variability in practice. The integration of computer models to help radiologists increase the accuracy of mammography examinations in diagnostic decision making has gained increasing attention in the last two decades. In this study, we provide an overview of one of the most commonly used models, artificial neural networks (ANNs), in mammography interpretation and diagnostic decision making and discuss important features in mammography interpretation. We conclude by discussing several common limitations of existing research on ANN-based detection and diagnostic models and provide possible future research directions.
    Computational and Mathematical Methods in Medicine 05/2013; 2013:832509. DOI:10.1155/2013/832509 · 0.77 Impact Factor
Show more