Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Potential Benefits of Computer-Aided Detection for Cancer Identification and Treatment—Reply

Potential Benefits of Computer-Aided Detection for Cancer Identification and Treatment—Reply In Reply Dr Brem believes our findings—that computer-aided detection (CAD) does not improve diagnostic accuracy of mammography—are curious and “incongruent with the overwhelming majority of peer-reviewed, published studies on the topic.” Her perspective is puzzling and ignores the large meta-analysis of 10 CAD studies which found that CAD significantly increased recall rates (and thus false-positive examinations) with no significant improvement in cancer detection rates compared with readings without CAD.1 Dr Brem selects 4 studies to support her perspective.2-5 Three studies, 2 from Europe2,3 and 1 from the United States,4 compared double reading to single reading with CAD. None measured single reading without CAD, and thus none were designed to evaluate the specific contribution of CAD to mammography performance. The single-site report4 from the United States by a single author that Dr Brem selects found no significant differences in sensitivity of single reading with CAD by a specialized mammographer compared with double reading by a generalist and a specialized mammographer.4 Dr Brem’s fourth reference is a retrospective cohort study5 of mammograms billed with and without CAD in a population of elderly women (80% aged 70-89 years and none younger than 67). In addition, no interpretations of any of the mammograms were available (the study was of Medicare claims data), and the authors could not control for confounding variables such as prior mammography, breast density, hormonal therapy, or radiologist. Fenton et al6 conclude from their study that diagnostic mammography, ultrasonography, and analysis of biopsy specimens were all significantly increased in this elderly population when CAD was used, and diagnoses of ductal carcinoma in situ increased significantly, as did surgery and potentially radiation in this elderly patient population. The benefits of these outcomes in women aged 70 to 89 years are unclear. More importantly, Dr Brem fails to cite the results and conclusions published subsequently by Fenton et al in their study6: “In conclusion, we found that, among large numbers of diverse facilities and radiologists, the use of computer software designed to improve the interpretation of mammograms was associated with significantly higher false positive rates, recall rates, and biopsy rates and with significantly lower overall accuracy in screening mammography than was nonuse.” As a practicing breast imager, I share Dr Brem’s passion to provide optimal care to patients. This optimal care includes providing high-quality examinations that add value and avoiding examinations that do not. As a researcher, I am equally passionate about the potential we have to extract more medical value from image data. The field of deep learning applied to image analysis is in its infancy and with rigorous and careful research will no doubt lead to new discoveries that add real value to image interpretation and our patients. Will rapid computing power and processing allow new frontiers in CAD and diagnosis? I would argue the potential is limitless. Are currently available commercial CAD products providing value to our patients? Careful scientific study has not found this to be true. Back to top Article Information Corresponding Author: Constance Lehman, MD, PhD, Department of Radiology, Massachusetts General Hospital, 55 Fruit St, Ste 240, Boston, MA 02114 (clehman@mgh.harvard.edu). Conflict of Interest Disclosures: Dr Lehman has received grant support from General Electric (GE) Healthcare and is a member of the Comparative Effectiveness Research Advisory Board for GE Healthcare. No other disclosures are reported. References 1. Taylor P, Potts HW. Computer aids and human second reading as interventions in screening mammography: two systematic reviews to compare effects on cancer detection and recall rate. Eur J Cancer. 2008;44(6):798-807.PubMedGoogle ScholarCrossref 2. Bargalló X, Santamaría G, Del Amo M, et al. Single reading with computer-aided detection performed by selected radiologists in a breast cancer screening program. Eur J Radiol. 2014;83(11):2019-2023.PubMedGoogle ScholarCrossref 3. Fenton JJ, Xing G, Elmore JG, et al. Short-term outcomes of screening mammography using computer-aided detection: a population-based study of Medicare enrollees. Ann Intern Med. 2013;158(8):580-587.PubMedGoogle ScholarCrossref 4. Gromet M. Comparison of computer-aided detection to double reading of screening mammograms: review of 231,221 mammograms. AJR Am J Roentgenol. 2008;190(4):854-859.PubMedGoogle ScholarCrossref 5. Gilbert FJ, Astley SM, Gillan MG, et al; CADET II Group. Single reading with computer-aided detection for screening mammography. N Engl J Med. 2008;359(16):1675-1684.PubMedGoogle ScholarCrossref 6. Fenton JJ, Taplin SH, Carney PA, et al. Influence of computer-aided detection on performance of screening mammography. N Engl J Med. 2007;356(14):1399-1409.PubMedGoogle ScholarCrossref http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JAMA Internal Medicine American Medical Association

Potential Benefits of Computer-Aided Detection for Cancer Identification and Treatment—Reply

JAMA Internal Medicine , Volume 176 (3) – Mar 1, 2016

Loading next page...
 
/lp/american-medical-association/potential-benefits-of-computer-aided-detection-for-cancer-36z1U0010i

References (7)

Publisher
American Medical Association
Copyright
Copyright © 2016 American Medical Association. All Rights Reserved.
ISSN
2168-6106
eISSN
2168-6114
DOI
10.1001/jamainternmed.2015.8474
Publisher site
See Article on Publisher Site

Abstract

In Reply Dr Brem believes our findings—that computer-aided detection (CAD) does not improve diagnostic accuracy of mammography—are curious and “incongruent with the overwhelming majority of peer-reviewed, published studies on the topic.” Her perspective is puzzling and ignores the large meta-analysis of 10 CAD studies which found that CAD significantly increased recall rates (and thus false-positive examinations) with no significant improvement in cancer detection rates compared with readings without CAD.1 Dr Brem selects 4 studies to support her perspective.2-5 Three studies, 2 from Europe2,3 and 1 from the United States,4 compared double reading to single reading with CAD. None measured single reading without CAD, and thus none were designed to evaluate the specific contribution of CAD to mammography performance. The single-site report4 from the United States by a single author that Dr Brem selects found no significant differences in sensitivity of single reading with CAD by a specialized mammographer compared with double reading by a generalist and a specialized mammographer.4 Dr Brem’s fourth reference is a retrospective cohort study5 of mammograms billed with and without CAD in a population of elderly women (80% aged 70-89 years and none younger than 67). In addition, no interpretations of any of the mammograms were available (the study was of Medicare claims data), and the authors could not control for confounding variables such as prior mammography, breast density, hormonal therapy, or radiologist. Fenton et al6 conclude from their study that diagnostic mammography, ultrasonography, and analysis of biopsy specimens were all significantly increased in this elderly population when CAD was used, and diagnoses of ductal carcinoma in situ increased significantly, as did surgery and potentially radiation in this elderly patient population. The benefits of these outcomes in women aged 70 to 89 years are unclear. More importantly, Dr Brem fails to cite the results and conclusions published subsequently by Fenton et al in their study6: “In conclusion, we found that, among large numbers of diverse facilities and radiologists, the use of computer software designed to improve the interpretation of mammograms was associated with significantly higher false positive rates, recall rates, and biopsy rates and with significantly lower overall accuracy in screening mammography than was nonuse.” As a practicing breast imager, I share Dr Brem’s passion to provide optimal care to patients. This optimal care includes providing high-quality examinations that add value and avoiding examinations that do not. As a researcher, I am equally passionate about the potential we have to extract more medical value from image data. The field of deep learning applied to image analysis is in its infancy and with rigorous and careful research will no doubt lead to new discoveries that add real value to image interpretation and our patients. Will rapid computing power and processing allow new frontiers in CAD and diagnosis? I would argue the potential is limitless. Are currently available commercial CAD products providing value to our patients? Careful scientific study has not found this to be true. Back to top Article Information Corresponding Author: Constance Lehman, MD, PhD, Department of Radiology, Massachusetts General Hospital, 55 Fruit St, Ste 240, Boston, MA 02114 (clehman@mgh.harvard.edu). Conflict of Interest Disclosures: Dr Lehman has received grant support from General Electric (GE) Healthcare and is a member of the Comparative Effectiveness Research Advisory Board for GE Healthcare. No other disclosures are reported. References 1. Taylor P, Potts HW. Computer aids and human second reading as interventions in screening mammography: two systematic reviews to compare effects on cancer detection and recall rate. Eur J Cancer. 2008;44(6):798-807.PubMedGoogle ScholarCrossref 2. Bargalló X, Santamaría G, Del Amo M, et al. Single reading with computer-aided detection performed by selected radiologists in a breast cancer screening program. Eur J Radiol. 2014;83(11):2019-2023.PubMedGoogle ScholarCrossref 3. Fenton JJ, Xing G, Elmore JG, et al. Short-term outcomes of screening mammography using computer-aided detection: a population-based study of Medicare enrollees. Ann Intern Med. 2013;158(8):580-587.PubMedGoogle ScholarCrossref 4. Gromet M. Comparison of computer-aided detection to double reading of screening mammograms: review of 231,221 mammograms. AJR Am J Roentgenol. 2008;190(4):854-859.PubMedGoogle ScholarCrossref 5. Gilbert FJ, Astley SM, Gillan MG, et al; CADET II Group. Single reading with computer-aided detection for screening mammography. N Engl J Med. 2008;359(16):1675-1684.PubMedGoogle ScholarCrossref 6. Fenton JJ, Taplin SH, Carney PA, et al. Influence of computer-aided detection on performance of screening mammography. N Engl J Med. 2007;356(14):1399-1409.PubMedGoogle ScholarCrossref

Journal

JAMA Internal MedicineAmerican Medical Association

Published: Mar 1, 2016

There are no references for this article.