System, method, and apparatus for temperature asymmetry measurement of body parts
Method for automatic detection of inflammation or functional disorder according to temperature asymmetry estimation in contralateral body parts is presented. Simultaneously recorded thermogram and the optical image of the inspected and contralateral body parts are sent to the processing unit, where they are stored, processed, and analyzed. Method automatically detects outlines of body parts in thermograms and optical images. Grid of points of interest is distributed inside the inspected and the contralateral body part's outline. Temperature maps are calculated according to both grids points and the temperature disparity map is estimated. The set of inflammation regions is obtained by analyzing the temperature disparity map and collecting adjacent points containing temperature differences surpassing the threshold. This is non-invasive and non-contact inspection method, suitable for real world environments with natural home or health care institutions background.
1 . A system for detection of a precursor to a diabetic foot ulcer based on temperature asymmetry estimation in feet, the system comprising:
a stand configured to rest on a surface and having a body part rest support configured to support feet;
an optical camera configured to detect optical image data corresponding to a first foot and optical image data corresponding to a second foot;
a thermal camera configured to detect a thermogram corresponding to the first foot and a thermogram corresponding to the second foot; and
a processor configured to:
receive the optical image data and the thermograms corresponding to the first foot and to the second foot,
estimate points of interest for both the first foot and the second foot,
estimate temperature maps for both the first foot and the second foot based on corresponding points of interest in optical image data and the thermograms,
estimate a temperature disparity map by subtracting temperature values for corresponding points of interest in the temperature maps for the first foot and the second foot,
compare the estimated temperature differences in the temperature disparity map to a threshold to identify candidate ulcer regions,
reject non-confident candidate ulcer regions using historically accumulated data indicating that the candidate ulcer region is likely not an ulcer or a precursor to the ulcer based on the candidate ulcer region being too small, and
identify the precursor to the diabetic foot ulcer in the first foot based on the candidate ulcer regions remaining in the temperature disparity map with non-confident candidate regions removed.
2 . The system of claim 1 , wherein the processor is further configured to estimate a recorded image displacement of the first foot based on the optical image data corresponding to the first foot and the thermogram corresponding to the first foot, to estimate a recorded image displacement of the second foot based on the optical image data corresponding to the second foot and the thermogram corresponding to the second foot, and to identify the precursor to the diabetic foot ulcer in the first foot based on the recorded image displacement of the first foot and based on the recorded image displacement of the second foot.
3 . The system of claim 2 , wherein estimating the recorded image displacement of the first foot includes localizing a template of the first foot by locating corresponding pairs of keypoints in the optical image data corresponding to the first foot and in the thermogram corresponding to the first foot, and calculating a nonconformity error between the corresponding pairs of keypoints in the optical image data corresponding to the first foot and in the thermogram corresponding to the first foot.
4 . The system of claim 1 , wherein the optical camera and the thermal camera are located on a mobile device, and the stand is configured to support the mobile device to direct the optical camera and the thermal camera towards the body part rest support.
5 . The system of claim 4 , wherein the mobile device further includes a network access device configured to transmit the thermograms and the optical image data to the processor.
6 . The system of claim 5 , wherein the mobile device further includes a display configured to output data indicating the identification of the precursor to the diabetic foot ulcer.
7 . A system for detection of a precursor to a diabetic foot ulcer based on temperature asymmetry estimation in feet, the system comprising:
an optical camera configured to detect optical image data corresponding to a first foot and optical image data corresponding to a second foot;
a thermal camera configured to detect a thermogram corresponding to the first foot and a thermogram corresponding to the second foot; and
a processor configured to:
receive the optical image data and the thermograms corresponding to the first foot and to the second foot,
estimate points of interest for both the first foot and the second foot,
estimate temperature maps for both the first foot and the second foot based on corresponding points of interest in optical image data and the thermograms,
estimate a temperature disparity map by subtracting temperature values for corresponding points of interest in the temperature maps for the first foot and the second foot,
compare the estimated temperature differences in the temperature disparity map to a threshold to identify candidate ulcer regions,
reject non-confident candidate ulcer regions using historically accumulated data indicating that the candidate ulcer region is likely not an ulcer or a precursor to the ulcer based on the candidate ulcer region being too small, and
identify the precursor to the diabetic foot ulcer in the first foot based on the candidate ulcer regions remaining in the temperature disparity map with non-confident candidate regions removed.
8 . The system of claim 7 , further comprising a body part rest support configured to support the first foot and the second foot.
9 . The system of claim 8 , further comprising a stand configured to rest on a surface and having the body part rest support coupled thereto.
10 . The system of claim 9 , wherein the stand further includes at least one camera holder configured to support the optical camera and the thermal camera relative to the body part rest support.
11 . The system of claim 10 , wherein the at least one camera holder includes a mobile device holder configured to support a mobile device, and wherein at least one of the optical camera or the thermal camera are located on the mobile device.
12 . The system of claim 11 , wherein the processor is at least one of located on the mobile device or a remote processor located remotely from the mobile device.
13 . The system of claim 10 , wherein the stand has a first end coupled to the body part rest support and a second end coupled to the at least one camera holder.
14 . The system of claim 9 , wherein the optical camera and the thermal camera are located on a mobile device, and the stand is configured to support the mobile device to direct the optical camera and the thermal camera towards the body part rest support.
15 . The system of claim 14 , wherein the mobile device further includes a network access device configured to transmit the thermogram and the optical image to the processor.
16 . The system of claim 15 , wherein the mobile device further includes a display configured to output data indicating the identification of the precursor to the diabetic foot ulcer.
17 . The system of claim 7 , wherein the processor is further configured to estimate a recorded image displacement of the first foot based on the optical image data corresponding to the first foot and the thermogram corresponding to the first foot, to estimate a recorded image displacement of the second foot based on the optical image data corresponding to the second foot and the thermogram corresponding to the second foot.
18 . The system of claim 17 , wherein estimating the recorded image displacement of the first foot includes localizing a template of the first foot by locating corresponding pairs of keypoints in the optical image data corresponding to the first foot and in the thermogram corresponding to the first foot, and calculating a nonconformity error between the corresponding pairs of keypoints in the optical image data corresponding to the first foot and in the thermogram corresponding to the first foot.
19 . A system for detection of a precursor to a diabetic foot ulcer based on temperature asymmetry estimation in feet, the system comprising:
an optical camera configured to detect optical image data corresponding to a first foot and optical image data corresponding to a second foot;
a thermal camera configured to detect a thermogram corresponding to the first foot and a thermogram corresponding to the second foot; and
a processor configured to:
receive the optical image data corresponding to the first foot and the optical image data corresponding to the second foot and to receive the thermogram corresponding to the first foot and the thermogram corresponding to the second foot,
estimate points of interest for both the first foot and the second foot,
estimate temperature maps for both the first foot and the second foot based on corresponding points of interest in optical image data and the thermograms,
estimate a temperature disparity map by subtracting temperature values for corresponding points of interest in the temperature maps for the first foot and the second foot,
compare the estimated temperature differences in the temperature disparity map to a threshold to identify candidate ulcer regions,
reject non-confident candidate ulcer regions using historically accumulated data indicating that the candidate ulcer region is likely not an ulcer or a precursor to the ulcer based on the candidate ulcer region being too small, and
identify the precursor to the diabetic foot ulcer in the first foot based on the candidate ulcer regions remaining in the temperature disparity map with non-confident candidate regions removed.
20 . The system of claim 7 wherein the processor is further configured to reject non-confident the candidate ulcer regions using historically accumulated data indicating that the candidate ulcer region is likely not the ulcer or the precursor to the ulcer due to the temperature difference in the temperature disparity map being too small.