Detection and characterization of cancerous tumors
Processes and techniques are disclosed to identify regions of suspected malignancy and their localization within a body part or organ. These methods rely on the analysis of infrared images to identify thermal abnormalities using image post-processing techniques, numerical modeling, iterative solutions methodology or a digital library. The methods utilize noninvasive, non-radiative and no contact infrared imaging that can be used for breast cancer screening for improved prognosis.
1. A method for identifying suspected malignancy within tissue using thermal imaging comprising:
a. constructing a geometric map of the surface of a body part;
b. generating thermal indicators from a thermal infrared image of the body part surface;
c. identifying a region of interest from the thermal indicators; and
d. identifying suspected malignancy within the body part, wherein the thermal indicators within the region of interest identifying malignancy comprises thermal abnormalities indicative of the presence of a tumor if the temperature difference between maximum and minimum temperatures is greater than 0.5° C. over a 5 mm to 40 mm range and thermal abnormalities are classified as veins if the temperature difference between maximum and minimum temperatures is between 0.2° C. and 2.0° C. over a 1 mm to 5 mm range.
2. The method of claim 1 , wherein the range indicative of the presence of a tumor is from 10 mm to 30 mm.
3. The method of claim 1 , wherein an aspect ratio of length of the largest dimension to width of the smallest dimension of the region of interest between 1 and 4 is indicative of malignancy and an aspect ratio greater than 5 is indicative of a blood vessel.
4. A method for determining the presence of heat generating tissue in a body part, comprising:
obtaining multi-view images of a body part, wherein the multi-view images comprise infrared images depicting a temperature distribution at the surface of the body part;
obtaining a 3-D digital model representing the surface and internal tissue of the body part;
generating a 3-D phantom thermal model by assigning values to selected parameters in the 3-D digital model of the body part and the surroundings and solving heat transfer equations, wherein the 3-D phantom thermal model comprises phantom thermal images of the surface of the body part;
comparing the phantom thermal images of the surface of the body part with the infrared images depicting a temperature distribution at the surface of the body part; and
identifying the values of the selected parameters generating the phantom thermal images matching the infrared thermal images.
5. The method of claim 4 , wherein the parameters include thermal conductivity of healthy tissue, thermal conductivity of heat generating tissue, metabolic activity of healthy tissue, metabolic activity of heat generating tissue, perfusion rate of healthy tissue, perfusion rate of heat generating tissue, perfusion rate in an area affected by heat generating tissue due to increased vasculature from angiogenesis, temperature of tissue at selected locations, heat transfer coefficient, heat transfer boundary conditions, emissivity of skin surface, heat transfer parameters representing a surface of the body part and surroundings, size and shape of a heat generating tissue mass, 3-D location of a heat generating tissue mass, or combinations thereof.
6. The method of claim 4 , wherein the selected parameters indicate a suspected malignancy.
7. The method of claim 4 , wherein the 3-D digital model is obtained from the multi-view images.
8. The method of claim 4 , wherein the 3-D digital model is obtained from a library of 3-D digital models representative of the body part.
9. The method of claim 4 , wherein the matching is performed by an iterative process.
10. The method of claim 4 , wherein the 3-D phantom thermal model is obtained from a library of 3-D phantom thermal models.
11. The method of claim 4 , wherein the body part is a female breast.
12. The method of claim 4 , wherein the multi-view images of the body part are obtained from an individual disposed in a prone position.
13. The method of claim 12 , wherein the body part is a female breast, further comprising hanging both breasts freely from a table and pulling one breast out of the way with a separating fabric to obtain a clear view of the other breast being imaged.
14. The method of claim 4 , wherein the 3-D digital model of the body part is generated from infrared, optical, ultrasound, magnetic resonance, outline capture techniques, shadow techniques, or X-ray multi-view images.
15. The method of claim 4 , wherein the matching is conducted by iteratively improving the match between the surface infrared images and the phantom thermal images until a defined convergence criterion is met.
16. The method of claim 4 , wherein the surface infrared images and the phantom thermal images are aligned using image registration in order to compare the surface infrared images and the generated phantom thermal images.
17. The method of claim 11 , wherein various geometric identifiers comprising breast shape, breast size, breast circumference, distance from chest wall to nipple, or volume are utilized in obtaining the 3-D digital model from the library.