Department of Biomedical Engineering Posters and Presentations
Breast Cancer Thermography: A Literature Review
Poster Number
106
Document Type
Poster
Publication Date
3-2016
Abstract
The current gold standard for breast cancer screening and detection is mammography. Approximately one-third of all biopsies ordered on the basis of mammography are negative and so women face unnecessary discomfort, anxiety, and occasional disfigurement. Breast thermography uses no radiation, does not require contact with the breast, and is more affordable. It has not yet, however, been shown to be a valid alternative or adjunct to mammography. This review provides a critical review of the work that has been done between 1979 and 2015 in the area of thermal imaging for breast cancer detection. Our intent is to guide and stimulate others interested in the field and to bring attention to the absence of standard scientific rigor in thermal imaging and its interpretation.
Forty papers were analyzed and categorized as segmentation (12), classification systems (4), system evaluation (12), data collection enhancement (4), or review (6). The review identifies the systems’ performance (sensitivity, specificity), and researchers’ analyses of results. This review highlights the deficiencies in the work done in the field and demonstrates that many research studies lack standardization. The review also describes the hardware used, images analyzed, and the results presented. It is clear that much of the research data and results are not reported comparably across studies. Sensitivity and specificity were presented in 16 of the 40 papers. Some papers were compared using their receiver operating characteristic curves. It plots true positive rate (sensitivity) against the false positive rate (1-sensitivity). Some performed only slightly better than chance (true positive rates and false positive rates are nearly equal). By reviewing these studies, others interested in the field of breast thermography will know what was done in the past, as well as what was and was not useful. Our lab is currently preparing to go into the clinic to take data for our own breast thermography project to test methods we have developed that will automate the detection process. A combination of high spatial and thermal resolution and the use of a subject as her own control may make it possible to perform accurate, objective, repeatable, and rapid thermal imaging as an adjunctive breast-imaging modality.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Open Access
1
Breast Cancer Thermography: A Literature Review
The current gold standard for breast cancer screening and detection is mammography. Approximately one-third of all biopsies ordered on the basis of mammography are negative and so women face unnecessary discomfort, anxiety, and occasional disfigurement. Breast thermography uses no radiation, does not require contact with the breast, and is more affordable. It has not yet, however, been shown to be a valid alternative or adjunct to mammography. This review provides a critical review of the work that has been done between 1979 and 2015 in the area of thermal imaging for breast cancer detection. Our intent is to guide and stimulate others interested in the field and to bring attention to the absence of standard scientific rigor in thermal imaging and its interpretation.
Forty papers were analyzed and categorized as segmentation (12), classification systems (4), system evaluation (12), data collection enhancement (4), or review (6). The review identifies the systems’ performance (sensitivity, specificity), and researchers’ analyses of results. This review highlights the deficiencies in the work done in the field and demonstrates that many research studies lack standardization. The review also describes the hardware used, images analyzed, and the results presented. It is clear that much of the research data and results are not reported comparably across studies. Sensitivity and specificity were presented in 16 of the 40 papers. Some papers were compared using their receiver operating characteristic curves. It plots true positive rate (sensitivity) against the false positive rate (1-sensitivity). Some performed only slightly better than chance (true positive rates and false positive rates are nearly equal). By reviewing these studies, others interested in the field of breast thermography will know what was done in the past, as well as what was and was not useful. Our lab is currently preparing to go into the clinic to take data for our own breast thermography project to test methods we have developed that will automate the detection process. A combination of high spatial and thermal resolution and the use of a subject as her own control may make it possible to perform accurate, objective, repeatable, and rapid thermal imaging as an adjunctive breast-imaging modality.
Comments
Presented at: GW Research Days 2016