Image analysis of digitized infrared images of the breasts from a first generation infrared imaging system

• Main Authors: Head, J.F., Lipari, C.A., Fen Wang, Elliott, R.L.
• Format: Conference Proceeding
• Language: English
• Subjects: Biomedical imaging; Breast cancer; Cancer detection; Image analysis; Infrared detectors; Infrared imaging; Medical diagnostic imaging; Pattern analysis; Risk management; Standardization
• ISSN: 1094-687X; 1558-4615
• DOI: 10.1109/IEMBS.1997.757718

Infrared imaging, often called thermography in medicine, of the breasts has been shown to be of value in risk assessment, detection, diagnosis and prognosis of breast cancer. However, infrared imaging has not been widely accepted for a variety of reasons, including the lack of standardization of the subjective visual analysis method. The subjective nature of the standard analysis makes it difficult to achieve equivalent results with different equipment and different interpreters of the infrared patterns of the breasts. Therefore, this study was undertaken to develop more objective methods of analysis of infrared images of the breasts by creating objective semiquantitative and quantitative analysis of computer assisted image analysis determined mean temperatures of whole breasts and quadrants of the breasts. When using objective quantitative data on whole breasts (comparing differences in means of left and right breasts), semiquantitative data on quadrants of the breast (determining an index by summation of scores for each quadrant), or summation of quantitative data on quadrants of the breasts there was a decrease in the number of abnormal patterns (positives) in patients being screen for breast cancer and an increases in the number of true positives in the breast cancer patients. It is hoped that the decrease in positives in women being screened for breast cancer will translate into a decrease in the false positives but larger numbers of women with longer follow-up will be needed to clarify this. Also a much larger group of breast cancer patients will need to be studied in order to see if there is a true increase in the percentage of breast cancer patients presenting with abnormal infrared images of the breast with these objective image analysis methods.