Method for analysis of pain images

• US 7,374,536 B1
• Filed: 04/16/2004
• Issued: 05/20/2008
• Est. Priority Date: 04/16/2004
• Status: Expired due to Fees

• Current Assignee: Taylor Valve Technology Incorporated
• Original Assignee: Taylor Valve Technology Incorporated
• Inventors: Taylor, Colin R.
• Primary Examiner(s): Winakur; Eric F.
• Assistant Examiner(s): ROZANSKI, MICHAEL T
• Application Number:
  US10/826,210
• Time in Patent Office: 1,495 Days
• Field of Search: 600/300, 600/424, 600/427, 600/553, 382/128, 382/266, 382/293, 607 59- 60, 702/19
• US Class Current: 600/300
• CPC Class Codes: A61B 5/00 Measuring for diagnostic purposes radiation diagnosis A61B6/00; diagnosis by ultrasonic, sonic or infrasonic waves A61B8/00; Identification of persons
• CPC Class Codes: A61B 5/4528 Joints A61B5/4533, A61B5/4538 take precedence
• CPC Class Codes: A61B 5/4824 Touch or pain perception evaluation A61B5/4821 takes precedence
• CPC Class Codes: A61B 5/743 Displaying an image simultaneously with additional graphical information, e.g. symbols, charts, function plots

Abstract

A method uses body images and computer hardware and software to collect and analyze clinical data in patients experiencing pain. Pain location information is obtained by the drawing of an outline of the pain on a paper copy or electronic display of the body image. Composite images are generated representing aggregate data for specified patient groups. The coordinates of common anatomic landmarks on differently designed body images are mapped to each other, permitting integrated analysis of pain data, e.g., pain shape, centroid, meta centroid, from multiple body image designs and display of all pain data on a single body image design. Differences and similarities between groups of patients are displayed visually and numerically, and are used to assign the probability of a given patient belonging to a particular diagnostic group or category of disease severity.

Claims

1. A method for identifying clinical characteristics of a patient'"'"'s pain, comprising the steps of:

a. providing a first human body image without pain indicia recorded thereon;

b. providing a second human body image substantially similar to said first human body image other than having recorded thereon readable indicia of a first patient'"'"'s pain, said indicia being recorded on said second human body image at locations in correspondence with locations of the first patient'"'"'s actual pain;

c. digitizing said first human body image and said second human body image if not already digitized;

d. coregistering said first human body image and said second human body image;

e. comparing electronically said first human body image and said second human body image to determine at least one set of contiguous electronic locations of said pain indicia;

f. calculating a single point centroid based on locations of actual pain, said actual pain locations comprising a pain shape;

g 1. using said single point centroid to allocate said pain shape to a given human body location;

g 2. producing a database of said pain centroids for a plurality of patients known to belong to at least one clinical group identified by at least one of a diagnostic category or a disease severity category;

h. comparing the first patient'"'"'s pain centroid with said database; and

i. judging based on said comparison if the first patient is included in at least one of said clinical groups.

2. A method as in claim 1 wherein said second human body image is created using a digitized version of said first human body image, said first human body image being electronically displayed for electronic recording of said pain indicia by a user.

3. A method as in claim 1, further comprising:

i. determining a pain shape determined by the outline of said set of contiguous electronic locations, wherein a pain shape for the first patient is matched with at least one pain shape pattern characteristic of said at least one clinical group, and comprising the further steps of;

repeating steps a—

g2 and i on a plurality of patients in said at least one clinical group to create a respective first visual display derived from pain shapes characteristic of said clinical group;

providing a first copy of said first human body image to which said first visual display is added;

reviewing the first patient'"'"'s copy of step b; and

identifying if said first visual display, characteristic of said at least one clinical group, corresponds to the location of pain provided by the first patient in step b.

4. A method as in claim 3 wherein a visual display of a pain shape is combined with other coded clinical information, comprising the further steps of:

determining a coding system for a second visual display of clinical information related to said pain shape and the first patient;

drawing by computer on a second copy of said digitized first human body image said pain shape for the first patient; and

drawing by computer on said second copy said second visual display that is visually linked to said pain shape for the first patient, whereby all relevant clinical information is combined in a single integrated display.

5. A method as in claim 3 wherein composite images are generated for a plurality of patients, comprising the steps of:

determining body locations of a plurality of said pain shapes;

determining a coding system of visual markers to provide information on said body locations of said pain shapes; and

providing a composite image in which said visual markers are added to a fourth copy of said first human body image so as to indicate the overall body location distribution of said pain shapes.

6. A method as in claim 5 wherein a smoothing of the composite image is obtained using parametric statistics, comprising the steps of:

calculating a meta-centroid for said pain shapes;

calculating a distance between said meta-centroid and the outline of each said pain shape along each radial from the meta-centroid outwards in one direction, disregarding both that portion of said outline that is on the other side of the meta-centroid and that portion of said outline that is the inner portion of the outline for an outline that does not encompass the meta-centroid;

calculating parametric statistics for said pain shapes for the aggregate of said distances for each said radial; and

providing a visual display on a fifth copy of said first human body image that shows the statistical likelihood, on the basis of parametric statistics, of a said pain shape occupying a given pixel at a given distance from the meta-centroid.

7. A method as in claim 3, wherein body-area-specific masks are used to focus body image analysis on particular body locations, comprising the further steps of:

providing at least one digitally marked body area on a third copy of said first human body image;

comparing said digitally marked body area and said centroid position and categorizing said pain shape as a member of the class of pains in said body area if said centroid lies inside said digitally marked body area.

8. A method as in claim 3, wherein said coregistering is achieved using anatomic landmarks identified by visual inspection of a human body and also landmarks that are identified only by palpation of a human body, comprising the further steps of:

providing said first human body image having both visual and palpable types of anatomic landmarks; and

visually displaying a coding system containing instructions for identification of at least one said anatomic landmark.

9. A method as in claim 3, wherein differentiation between said first human body image and said second human body image is obtained by using different colors that can be separately identified by computer, comprising the further steps of:

producing a said first human body image in which the body outline and other anatomic markings are printed in a first color that has a low value for at least one of three constituent RGB colors that is easily visible to a human being;

recording said readable indicia on said second human body image using a marking device that produces markings in a second color that has a greater value than with the first color for at least one of said constituent RGB colors;

producing a computer-readable color digitized version of said second human body image; and

producing a digitized image confined to said readable indicia by having said computer select pixels for analysis based on differences in values between said constituent RGB colors in said color digitized version.

10. A method as in claim 3, wherein said pain indicia that have been improperly recorded are edited prior to identification of said electronic locations, comprising the further steps of:

analyzing said pain indicia on said second human body image to identify non-contiguous pain indicia in which all pixels are not contiguous;

identifying which of said non-contiguous pain indicia may properly be considered editable pain indicia; and

incorporating intervening pixels between said editable pain indicia with said editable pain indicia so as to form a single pain indicia object containing a contiguous set of pixels.

11. A method as in claim 1 wherein body-area-specific masks are used to focus body image analysis on particular body locations, comprising the further steps of:

providing at least one digitally marked body area on a third copy of said first human body image;

comparing said digitally marked body area and said centroid position and categorizing said pain shape as a member of the class of pains in said body area if said centroid lies inside said digitally marked body area.

12. A method as in claim 1 wherein said coregistering is achieved using anatomic landmarks identified by visual inspection of a human body and also landmarks that are identified only by palpation of a human body, comprising the further steps of:

providing said first human body image having both visual and palpable types of anatomic landmarks; and

visually displaying a coding system containing instructions for identification of at least one said anatomic landmark.

13. A method as in claim 1 wherein differentiation between said first human body image and said second human body image is obtained by using different colors that can be separately identified by computer, comprising the further steps of:

producing a said first human body image in which the body outline and other anatomic markings are printed in a first color that has a low value for at least one of three constituent RGB colors that is easily visible to a human being;

recording said readable indicia on said second human body image using a marking device that produces markings in a second color that has a greater value than with the first color for at least one of said constituent RGB colors;

producing a computer-readable color digitized version of said second human body image; and

producing a digitized image, confined to said readable indicia by having said computer select pixels for analysis based on differences in values between said constituent RGB colors in, said color digitized version.

14. A method as in claim 1 wherein said pain indicia that have been improperly recorded are edited prior to identification of said electronic locations comprising the further steps of:

analyzing said pain indicia on said second human body image to identify non-contiguous pain indicia in which all pixels are not contiguous;

identifying which of said non-contiguous pain indicia may properly be considered editable pain indicia; and

incorporating intervening pixels between said editable pain indicia with said editable pain, indicia so as to form, a single pain indicia object containing a contiguous set of pixels.

15. A method as in claim 1 wherein step h is performed by incorporating information on both pain location and pain quality.

16. A method as in claim 1 wherein human body image and said second human body image are based on a photograph of the first patient.