Department of Biomedical Engineering Posters and Presentations
Development of a Breast Thermogram Analysis Tool
Poster Number
111
Document Type
Poster
Publication Date
3-2016
Abstract
Although the gold standard for breast cancer screening and evaluation is mammography, we continue to seek complementary modalities that may reduce or eliminate its requirements for radiation and physical compression. Tumor growth causes angiogenesis, which is the growth of new blood vessels to supply the tumor. The consequent increased blood flow causes the temperature to increase around the tumor and on the surface of the breast. Passive, non-contact infrared measurement of the temperature (thermography) would seem, therefore, to be a good candidate as a complementary modality for screening and/or diagnosis. Previous theoretical and simulation studies in our lab have shown that certain relationships exist between the presences of a tumor and increased localized surface temperature [1]. We are now developing a system for automated abnormality detection and analysis using thermal images of breasts. An automatic segmentation method is used to extract the breast regions from the acquired images; this limits the area to be searched for locally-warm regions. A mapping program then is used to find the region on the contralateral breast that corresponds to the ipsilateral high-temperature region. By comparing these two corresponding regions, we can decide whether the candidate abnormal region has a locally or global higher temperature. The segmentation and mapping tools have been validated with real and simulated data. A pilot study is still under review by the IRB committee, but the development of this system will benefit and speed up both the collection and analysis of the thermal images soon to be acquired.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Open Access
1
Development of a Breast Thermogram Analysis Tool
Although the gold standard for breast cancer screening and evaluation is mammography, we continue to seek complementary modalities that may reduce or eliminate its requirements for radiation and physical compression. Tumor growth causes angiogenesis, which is the growth of new blood vessels to supply the tumor. The consequent increased blood flow causes the temperature to increase around the tumor and on the surface of the breast. Passive, non-contact infrared measurement of the temperature (thermography) would seem, therefore, to be a good candidate as a complementary modality for screening and/or diagnosis. Previous theoretical and simulation studies in our lab have shown that certain relationships exist between the presences of a tumor and increased localized surface temperature [1]. We are now developing a system for automated abnormality detection and analysis using thermal images of breasts. An automatic segmentation method is used to extract the breast regions from the acquired images; this limits the area to be searched for locally-warm regions. A mapping program then is used to find the region on the contralateral breast that corresponds to the ipsilateral high-temperature region. By comparing these two corresponding regions, we can decide whether the candidate abnormal region has a locally or global higher temperature. The segmentation and mapping tools have been validated with real and simulated data. A pilot study is still under review by the IRB committee, but the development of this system will benefit and speed up both the collection and analysis of the thermal images soon to be acquired.
Comments
Presented at: GW Research Days 2016