Detecting tumorous breast tissue in a thermal image
What is disclosed is a system and method for detecting cancerous tissue in breast tissue using a thermal image. In one embodiment, the present tumor detection method involves selecting a region of interest in the thermal image to be processed for breast cancer screening. Thereafter, a percentage of pixels p 1 in the selected region having a temperature T pixel 1 , where T 1 ≦T pixel 1 ≦T 2 , is determined. A percentage of pixels p 2 in the selected region having a temperature T pixel 2 , where T 3 ≦T pixel 2 , is determined. A ratio p 3 = P edge / P block is also determined, where P edge is a number of pixels around a border of a suspected tumor within the selected region, and P block is a number of pixels in a perimeter of the selected region. A decision fusion rule R, as more fully disclosed herein, is utilized to determine, based on these determination whether tissue within that region is cancerous, or non-cancerous, or is suspicious of being cancerous.
1. A method for detecting tumorous breast tissue using a thermal image, the method comprising:
receiving a thermal image of breast skin tissue of a subject being screened for breast cancer, said thermal image having been captured by a thermal imaging system and comprising a plurality of pixels each having a corresponding temperature value;
selecting at least one region of interest in said thermal image;
determining a percentage of pixels p 1 in said selected region with a temperature T pixel 1 , where T 1 ≦T pixel 1 ≦T 2 ;
determining a ratio
p
3
=
P
edge
/
P
block
,
where P edge is a number of pixels around a border of a suspected tumor within said selected region, and P block is a number of pixels in a perimeter of said selected region; and
using a decision fusion rule R to determine that tissue in said region of interest is one of: cancerous, non-cancerous, or suspicious of being cancerous.
2. The method of claim 1 , wherein said thermal imaging system is any of: a single-band infrared camera, a multi-band infrared camera in the thermal range, and a hyperspectral infrared camera in the thermal range.
3. The method of claim 1 , wherein said decision fusion rule R is based on any combination of: R 1 ,R 2 ,R 3 , where: R 1 =(p 1 ≧Threshold 1 ), R 2 =(p 2 ≧Threshold 2 ), and R 3 =(p 3 ≧Threshold 3 ).
4. The method of claim 3 , wherein Threshold 1 , Threshold 2 , Threshold 3 comprise user-defined threshold values based on any combination of: tumor characteristics, subject age, race, sex, and medical history.
5. The method of claim 3 , where said decision fusion rule, R, comprises any of: a majority rule of (R 1 , R 2 , R 3 ), and a weighted sum of (R 1 , R 2 , R 3 ).
6. The method of claim 1 , wherein T 1 =T max , where T max is a maximum temperature of said subject, and T 2 =└T 1 ┘−1° C.
7. The method of claim 1 , wherein T 3 =T avg +(T max −T avg )/3, where T max is a maximum temperature of said subject, and T avg is an average temperature of said subject.
8. The method of claim 1 , wherein, in response to having performed screening, displaying a message indicating a probability that tissue in said region of interest can be categorized as being in any category of: thermal BIRADs category, thermo-biological category, non-cancerous category, suspicious of being cancerous category, and cancerous category.
9. The method of claim 1 , further comprising communicating any of said determinations to any of: a medical staff, a medical data repository, a memory, a storage device, a display device, a handheld wireless device, a handheld cellular device, and a remote device over a network.
10. The method of claim 1 , wherein, in response to a determination that tissue in said selected region is cancerous, performing any of: initiating an alert, and signaling a medical professional.
11. A system for detecting tumorous breast tissue using a thermal image, the system comprising:
a memory and a storage device; and
a microprocessor in communication with said memory and storage device, said microprocessor executing machine readable program instructions for performing:
receiving a thermal image of breast skin tissue of a subject being screened for breast cancer, said thermal image having been captured by a thermal imaging system and comprising a plurality of pixels each having a corresponding temperature value;
selecting at least one region of interest in said thermal image;
determining a percentage of pixels p 1 in said selected region having a first temperature T pixel 1 , where T 1 ≦ pixel 1 ≦T 2 ;
determining a ratio
p
3
=
P
edge
/
P
block
,
where P edge is a number of pixels around a border of a suspected tumor within said selected region, and P block is a number of pixels in a perimeter of said selected region; and
using a selected decision fusion rule, R, to determine that tissue within said region of interest is one of: cancerous, non-cancerous, or suspicious of being cancerous.
12. The system of claim 11 , wherein said thermal imaging system is any of: a single-band infrared camera, a multi-band infrared camera in the thermal range, and a hyperspectral infrared camera in the thermal range.
13. The system of claim 11 , wherein said decision fusion rule R is based on any combination of: R 1 ,R 2 ,R 3 , where: R 1 =(p 1 ≧Threshold 1 ), R 2 =(p 2 ≧Threshold 2 ), and R 3 =(p 3 ≧Threshold 3 ).
14. The system of claim 13 , wherein Threshold 1 , Threshold 2 , Threshold 3 comprise user-defined threshold values based on any combination of: tumor characteristics, subject age, race, sex, and medical history.
15. The system of claim 13 , where said decision fusion rule, R, comprises any of: a majority rule of (R 1 , R 2 , R 3 ), and a weighted sum of (R 1 , R 2 , R 3 ).
16. The system of claim 11 , wherein T 1 =T max , where T max is a maximum temperature of said subject, and T 2 =└T 1 ┘−1° C.
17. The system of claim 11 , wherein T 3 =T avg +(T max −T avg )/3, where T max is a maximum temperature of said subject, and T avg is an average temperature of said subject.
18. The system of claim 11 , wherein, in response to having performed screening, displaying a message indicating a probability that tissue in said region of interest can be categorized as being in any category of: thermal BIRADs category, thermo-biological category, non-cancerous category, suspicious of being cancerous category, and cancerous category.
19. The system of claim 11 , further comprising communicating any of said determinations to any of: a medical staff, a medical data repository, a memory, a storage device, a display device, a handheld wireless device, a handheld cellular device, and a remote device over a network.
20. The system of claim 11 , wherein, in response to a determination that tissue in said selected region is cancerous, performing any of: initiating an alert, and signaling a medical professional.