Single Reading with Computer-Aided Detection for Screening Mammography

• Published in: The New England journal of medicine Vol. 359; no. 16; pp. 1675 - 1684
• Main Authors: Gilbert, Fiona J, Astley, Susan M, Gillan, Maureen G.C, Agbaje, Olorunsola F, Wallis, Matthew G, James, Jonathan, Boggis, Caroline R.M, Duffy, Stephen W
• Format: Journal Article
• Language: English
• Published: Boston, MA Massachusetts Medical Society 16.10.2008
• Subjects: Aged; Biological and medical sciences; Breast - pathology; Breast cancer; Breast Neoplasms - diagnostic imaging; Breast Neoplasms - pathology; Clinical outcomes; Clinical trials; Female; General aspects; Genital system. Mammary gland; Humans; Information systems; Investigative techniques, diagnostic techniques (general aspects); Mammography; Mammography - methods; Medical sciences; Medical screening; Middle Aged; Predictive Value of Tests; Radiodiagnosis. Nmr imagery. Nmr spectrometry; Radiographic Image Interpretation, Computer-Assisted; Radiology; Sensitivity and Specificity; Human; Breast disease; Radiodiagnosis; Breast cancer; Mammography; Malignant tumor; Medical screening; Medicine; Mammary gland diseases; Reading; Female; Diagnosis; Detection; Woman; Computer aid; Cancer
• ISSN: 0028-4793; 1533-4406
• DOI: 10.1056/NEJMoa0803545

This large trial of methods used for the evaluation of screening mammograms compared the rate of detection of breast cancer achieved by the use of two readers for each mammogram with that of a single reader using computer-aided detection. The detection rates were virtually the same for both methods, but the recall rate was slightly but significantly higher for mammograms read by a single reader with computer-aided detection than for those read by two readers. This trial of methods used for the evaluation of screening mammograms compared the rate of detection of breast cancer achieved by the use of two readers for each mammogram with that of a single reader using computer-aided detection. The interpretation of screening mammograms by two readers (double reading) rather than by one reader increased the rate of cancer detection by 4 to 14% in several studies 1 – 6 and by 10% according to a meta-analysis. 7 Double reading is standard practice in at least 12 European countries 8 but has not been widely adopted in the United States. The use of computer-aided detection systems has the potential to increase cancer-detection rates. Thus, a single reader using computer-aided detection might be able to match the performance of two readers. 9 , 10 Single reading with computer-aided detection could be an alternative to double reading . . .