Method and system for non-invasive temperature mapping of tissue

• US 6,067,371 A
• Filed: 02/13/1998
• Issued: 05/23/2000
• Est. Priority Date: 11/28/1995
• Status: Expired due to Fees

• Current Assignee: Dornier Medical Systems, Inc. (Mercedes-Benz Group AG)
• Original Assignee: Dornier Medical Systems, Inc. (Mercedes-Benz Group AG)
• Inventors: Burdette, Everette C., diMonda, Richard, Gouge, James
• Primary Examiner(s): BELLA, MATTHEW C
• Application Number:
  US09/022,797
• Time in Patent Office: 830 Days
• Field of Search: 382/128, 382/157, 382/218, 600/412, 600/549
• US Class Current: 382/128
• CPC Class Codes: A61B 5/015 By temperature mapping of body part
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: A61B 8/5223 for extracting a diagnostic or physiological parameter from medical diagnostic data for algorithms to analyse biomedical images G06T7/0012
• CPC Class Codes: G01K 3/14 in respect of space
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment

Abstract

A method and system for mapping temperature from image data. The method includes the steps of: receiving an image of tissue comprised of multiple pixels, segregating the image into groups of pixels (104), each group of pixels having a set of descriptors (106), establishing a baseline set of descriptors corresponding to initial conditions of the imaged tissue (108), measuring a differential in the set of descriptors for a group of pixels (114), the differential corresponding to a change in pixel values for the group of pixels, correlating said measured differential to a temperature change for the tissue corresponding to the group of pixels (118-124), and overlaying an indication of temperature over the tissue image in response to said correlated temperature change indicating a change in temperature range for the tissue (130-132). The system includes a digital processor implementing a neural network for evaluating a set of descriptors against baseline values learned by the neural network form the initial conditions of the tissue.

Claims

1. A method for mapping temperature changes in tissue comprising:

receiving an image of tissue comprised of multiple pixels;

segregating said image into groups of pixels, each said group of pixels having a corresponding set of descriptors derived from said pixels in said corresponding group of pixels;

establishing a baseline for at least one of each group of pixels and said corresponding set of descriptors, said baseline corresponding to initial conditions of the tissue as a learning set;

measuring a differential between said set of descriptors for a group of pixels which correspond to new pixels within a group of pixels and said baseline for said set of descriptors for said group of pixels;

correlating said measured differential to a temperature change for the tissue corresponding to said group of pixels; and

overlaying an indication of temperature over the tissue image in response to said correlated temperature change indicating a change in temperature range for the tissue.

2. The method of claim 1, said segregating step further comprising the step of:

defining a region of interest in said image of tissue; and

segregating said region of interest into groups of pixels.

3. The method of claim 1, said segregating step further comprising the step of:

selecting said set of descriptors for each group of pixels from a set of descriptors formed by a differential measurement between pixels within said group and pixels outside said group.

4. The method of claim 1, said correlating step further comprising:

reading an actual temperature measurement corresponding to one of said groups of pixels;

measuring a grayscale differential between said baseline for said one group of pixels; and

correlating said grayscale differential to a difference between said actual temperature measurement and a baseline temperature measurement.

5. The method of claim 4 further comprising the step of:

adjusting a grayscale-to-temperature correlation factor for said other groups of pixels in correspondence with said correlated grayscale to temperature measurement.

6. The method of claim 1, said establishing baseline step further comprising:

calibrating a grayscale-to-temperature correlation factor from a plurality of power deposition measurements.

7. The method of claim 1, said modifying step further comprising the steps of:

generating overlay data corresponding to said predetermined temperature ranges, said overlay data comprising a color for each of said predetermined temperature ranges.

8. A system for mapping temperature changes in tissue comprising:

a buffer for receiving video data corresponding to an image generated for an ultrasound system;

a processor for establishing baseline values from video data corresponding to at least one region of said image and generating descriptors for said region derived from video data corresponding to said region, said video data for generating said descriptors being received subsequent to said video data used to establish said baselines;

a comparator for comparing said generated descriptors to said established baselines to measure image change;

a correlator to correlate said measured image change to a temperature range; and

an overlay generator for generating overlay image data for modifying said image to indicate said region being in said correlated temperature range.

9. The system of claim 8, said comparator and said correlator being implemented in a neural network of neural elements, each of said neural elements implementing a transfer function for correlating measured image change to a temperature range.

10. The system of claim 9, said neural elements further including:

input for temperature data about tissue corresponding to said image; and

said neural element modifying said transfer function in correspondence to said temperature data so that said transfer function more accurately correlates said measured image changes to said temperature range.

11. A method for mapping temperature changes in tissue comprising the steps of:

segregating an image of tissue comprised of multiple pixels into a target area of target pixels;

extracting pattern data from said target area in an initial image of the tissue to establish a baseline for each pattern in said pattern data extracted from said initial image;

extracting pattern data from said target area in a subsequent image of said tissue;

determining a best correspondence between said extracted pattern data in said subsequent image and said extracted pattern data in said initial image by comparing said extracted pattern data in said subsequent image with said pattern data in said initial image;

measuring a differential between said extracted pattern data from said subsequent image and said established baseline for said extracted pattern best corresponding to said pattern data in said initial image;

correlating a temperature change to said measured differential for said extracted pattern data in said subsequent image; and

overlaying an indication of temperature range over a portion of said target area corresponding to said correlated temperature change to indicate a change in temperature range.

12. The method of claim 11, said extracting pattern data steps further comprising the steps of:

extracting radial data about said target pixels within one of said initial image and said subsequent image; and

extracting line data for said target pixels within one of said initial image and said subsequent image.

13. The method of claim 12, said extracting radial data step further comprising the step of:

extracting for each target pixel radial shape data, radial spatial data, radial contour data, radial asymmetry data, and radial gradient data.

14. The method of claim 12 further comprising the step of extracting geometric shape data from said extracted radial and said extracted line data.