Method for controlling adaptive color flow processing using fuzzy logic

• US 5,357,965 A
• Filed: 11/24/1993
• Issued: 10/25/1994
• Est. Priority Date: 11/24/1993
• Status: Expired due to Fees

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Hall, Anne L., Harsh, Michael J.
• Primary Examiner(s): MANUEL, GEORGE C
• Application Number:
  US08/158,071
• Time in Patent Office: 335 Days
• Field of Search: 128/4, 128/661.01, 128/661.08, 128/661.09, 128/661.10, 128/719, 128/731, 364/413.25
• US Class Current: 600/454
• CPC Class Codes: G01S 15/8981 Discriminating between fixed and moving objects or between objects moving at different speeds, e.g. wall clutter filter
• CPC Class Codes: G01S 7/52071 Multicolour displays; using colour coding; Optimising colour or information content in displays, e.g. parametric imaging

Abstract

A color flow processor has a fuzzy logic processor for determining when an adaptive wall filter can be turned off in response to the condition wherein flow signal will be treated as wall signal. The fuzzy logic processor uses details of wall velocity and power, as well as variance, to determine whether the measured echo signal component to be filtered truly represents the wall velocity only. The general rule applied by the fuzzy logic processor would be that if the wall velocity is LOW and the wall variance is LOW and the wall power is HIGH, than the adaptive filter is turned ON, where LOW and HIGH are fuzzy values. Use can be made of information from previous states, either temporally or spatially, allowing the system to adapt itself to each study or over time.

Claims

1. In a processor for color flow imaging of the velocity of blood flowing past adjacent tissue, comprising a wall estimate path circuit and a flow estimate path, and having a nonadaptive mode in which said wall estimate path circuit is bypassed and an adaptive mode in which the estimated wall velocity of said adjacent tissue is mixed with I and Q input data in a complex mixer in said wall estimate path circuit and then output to a wall filter in said flow estimate path circuit, the improvement wherein a fuzzy logic processor in said wall estimate path circuit for determining which of said adaptive and nonadaptive modes shall be operative.

2. The color flow processor as defined in claim 1, wherein said wall estimate path circuit further comprises an autocorrelator connected to a wall velocity estimator for outputting a value corresponding to the estimated wall velocity to said complex mixer and to said fuzzy logic processor.

3. The color flow processor as defined in claim 2, wherein said fuzzy logic processor comprises first crisp-to-fuzzy transform means for mapping actual values of estimated wall velocity to a degree of membership in at least one wall velocity fuzzy set.

4. The color flow processor as defined in claim 2, wherein said wall estimate path circuit further comprises a turbulence estimator connected to receive a value corresponding to estimated wall power from said autocorrelator, to receive a value corresponding to estimated wall variance from said wall velocity estimator, and to output a value corresponding to estimated wall variance to said fuzzy logic processor.

5. The color flow processor as defined in claim 4, wherein said fuzzy logic processor comprises first crisp-to-fuzzy transform means for mapping actual values of estimated wall velocity to a degree of membership in at least one wall velocity fuzzy set, and second crisp-to-fuzzy transform means for mapping actual values of estimated wall variance to a degree of membership in at least one wall variance fuzzy set.

6. The color flow processor as defined in claim 4, wherein said fuzzy logic processor is connected to receive said value corresponding to estimated wall power from said autocorrelator.

7. The color flow processor as defined in claim 6, wherein said fuzzy logic processor comprises first crisp-to-fuzzy transform means for mapping actual values of estimated wall velocity to a degree of membership in at least one wall velocity fuzzy set, second crisp-to-fuzzy transform means for mapping actual values of estimated wall variance to a degree of membership in at least one wall variance fuzzy set, and third crisp-to-fuzzy transform means for mapping actual values of estimated wall power to a degree of membership in at least one wall power fuzzy set.

8. The color flow processor as defined in claim 7, wherein said fuzzy logic processor further comprises fuzzy-to-crisp transform means for outputting a wallness signal which controls whether estimated wall velocity values will be mixed with said I and Q input data in said complex mixer, said wallness signal being a function of said values corresponding to estimated wall velocity, variance and power.

9. The color flow processor as defined in claim 8, wherein said fuzzy logic processor further comprises inference means for outputting said wallness signal in accordance with a set of fuzzy logic rules.

10. The color flow processor as defined in claim 9, wherein said fuzzy logic rules include the rule that if the wall velocity is LOW and the wall variance is LOW and the wall power is HIGH, than wallness is HIGH, where LOW and HIGH are fuzzy values, said adaptive mode being turned ON in response to said HIGH wallness having a degree of membership greater than a predetermined threshold.

11. A method for color flow imaging of the velocity of blood flowing past adjacent tissue, comprising the steps of:

acquiring I and Q input data;

estimating the wall velocity of said adjacent tissue from said I and Q input data;

mixing said estimated wall velocity with said I and Q input data in response to said I and Q input data forming a set of data which is a member of a predetermined class of data sets;

mixing a zero wall velocity with said I and Q input data in response to said I and Q input data forming a set of data which is not a member of said predetermined class of data sets; and

determining whether said I and Q input data is a member of said predetermined class of data sets using fuzzy logic processing.

12. The method as defined in claim 11, wherein said fuzzy logic processing comprises the step of transforming said estimated wall velocity to a degree of membership in at least one wall velocity fuzzy set.

13. The method as defined in claim 11, further comprising the step of estimating the wall variance of said adjacent tissue from said I and Q input data, wherein said fuzzy logic processing comprises the steps of transforming said estimated wall velocity to a degree of membership in at least one wall velocity fuzzy set and transforming said estimated wall variance to a degree of membership in at least one wall variance fuzzy set.

14. The method as defined in claim 11, further comprising the steps of estimating the wall variance and wall power of said adjacent tissue from said I and Q input data, wherein said fuzzy logic processing comprises the steps of transforming said estimated wall velocity to a degree of membership in at least one wall velocity fuzzy set, transforming said estimated wall variance to a degree of membership in at least one wall variance fuzzy set and transforming said estimated wall power to a degree of membership in at least one wall power fuzzy set.

15. The method as defined in claim 14, wherein said fuzzy logic processing further comprises the steps of outputting a wallness signal which controls whether estimated wall velocity will be mixed with said I and Q input data, said wallness signal being output as a function of said values corresponding to estimated wall velocity, variance and power.

16. The method as defined in claim 15, wherein said fuzzy logic processing further comprises the step of outputting said wallness signal in accordance with a set of fuzzy logic rules.

17. The method as defined in claim 16, wherein said fuzzy logic rules include the rule that if the wall velocity is LOW and the wall variance is LOW and the wall power is HIGH, than wallness is HIGH, where LOW and HIGH are fuzzy values, said adaptive mode being turned ON in response to said HIGH wallness having a degree of membership greater than a predetermined threshold.