Using the Brazilian visual lab mastology data set containing dynamic thermograms, applying components analysis resulted in two corresponding images: 1. Slight decrease as function of rewarming (suggests correlation with the cancer generated heat), 2. Temperature increase as function of rewarming (suggests correlation with veins affected by vasomodulation). All components appear clear and distinct. Applying signal processing methods to the dynamic infrared data, we found distinct components with correspondence to understood physiologic processes. The cases shown are self-evident of the capability of the method but are lacking supporting ground truth that is unavailable with such a limited data set.

I suggest a method for biomedical imaging with heat using principal and independent components analysis. This method produces novel results suggesting physiologic mechanisms. When using thermal imaging to detect breast cancer, the dominant heat signature is of indirect heat transported by the blood away from the tumor location into the skin. Interpretation is usually based on vascular patterns and not by observing the direct cancerous heat. In this new method one uses a sequence of thermal images of the patient’s breast following external temperature change. Data are recorded and analyzed using independent component analysis (ICA) and principal component analysis (PCA). ICA separates the image sequence into new independent images having a common characteristic time behavior. Using the Brazilian visual lab mastology data set, I observed three types of component images: Images corresponding to a minimum change as a function of applied temperature or time, which suggests an association with the cancer generated heat, images in which a moderate temperature dependence is associated with veins affected by vasomodulation, and images of complex time behavior indicating heat absorption due to high perfusion of the tumor. All components appear to clearly and distinctly represent underlying physiology.