METHOD, SYSTEM AND DEVICE FOR ENHANCING FLOW FIELD DATA

• US 20110144920A1
• Filed: 08/25/2008
• Published: 06/16/2011
• Est. Priority Date: 08/25/2008
• Status: Active Grant

• Current Assignee: ETH Zurich, Indian Institute of Technology Kharagpur
• Original Assignee: ETH Zurich, Indian Institute of Technology Kharagpur
• Inventors: McGregor, Robert, Szczerba, Dominik, Muralidhar, Krishnamurthy, Szekely, Gabor
• Granted Patent: US 8,898,022 B2
• Time in Patent Office: Days
• US Class Current: 702/19
• CPC Class Codes: G06F 2111/10 Numerical modelling
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders

Abstract

A method, system and device are described for generating high-resolution flow-field data from sparse measurement data by interpolating from a reference database of pre-computed, highly-resolved fluid dynamic simulation results, to generate a highly-resolved set of flow field data which correspond to the sparse measurements. In one embodiment, the method, system and device use Principal Component Analysis techniques to interpolate the multidimensional fluid flow data. Applications of the invention are, amongst others, in the field of biomedical diagnosis involving fluid dynamic modeling of biological fluid systems such as blood flow in an artery.

Claims

1. Method of obtaining highly resolved flow field data representing the flow characteristics of fluid in a fluid system under analysis, comprising a first step of measuring and/or estimating parameters of the fluid system under analysis to produce a set of flow predictor parameters for the fluid system under analysis,

the method comprising the further steps of,

using a reference database containing highly resolved flow field data representing fluid flow simulation results of a reference set of fluid systems and flow predictor parameters of each of said set of reference fluid systems, performing a mathematical interpolation operation on said reference database to determine an interpolated set of flow predictor parameters, matching, to within predetermined error margins, the said set of flow predictor parameters of the fluid system under analysis, and

determining, from said reference database, a set of highly resolved flow field data corresponding to the said interpolated set of flow predictor parameters.

2. Method according to claim 1, including the steps of performing a vector transform operation on the highly resolved flow field data and the flow predictor parameters in the database, and performing the mathematical interpolation operation on the transformed data.

3. Method according to claim 2, in which the vector transform operation is a Principal Component Analysis, a Factor Analysis or an Independent Component Analysis.

4. Method according to claim 1, in which the fluid system under analysis is a biological fluid flowing in a biological conduit.

5. Method according to claim 4, in which the fluid system under analysis is blood flowing through a branching geometry of a blood vessel.

6. Method according to claim 1, further comprising a normalizing step for mapping the measured and/or estimated flow predictor parameters for the fluid systems observed on to a common coordinate system used in the reference database.

7. Method according to claim 1, in which the step of measuring and/or estimating fluid flow parameters of the fluid system under analysis includes the steps of measuring and/or estimating fluid flow parameters of the fluid system under analysis at each of a plurality of time intervals, and in which the set of flow predictor parameters for the fluid system under analysis includes flow predictor parameters for each of said time intervals.

12. Method of producing a reference set of high resolution flow field vector data suitable for the method of claim 1, comprising the steps of, for each of a plurality of reference fluid systems,

measuring and/or estimating fluid flow predictor parameters at each of a plurality of time intervals,

performing high-resolution flow simulations using the said fluid flow predictor parameters, so as to thereby produce high-resolution flow field data descriptive of fluid flow in each of said reference fluid systems, and

storing in a reference database the said fluid flow predictor parameters and said high-resolution flow field data produced by said high-resolution flow simulations.

13. Method according to claim 12, including the step of normalizing the fluid flow predictor parameters for the plurality of reference fluid systems so as to map the fluid flow predictor parameters on to a common set of coordinates at each of a plurality of time intervals.

14. Method according to claim 12, in which the step of storing in a reference database the said fluid flow predictor parameters and said high-resolution flow field data includes the step of performing a mathematical transform on the said fluid flow predictor parameters and the said high-resolution flow field data, such that the fluid flow predictor parameters and high-resolution flow field data are stored in the reference database as the results of the mathematical transform.

15. Method according to claim 12, in which the vector transform operation is a Principal Component Analysis, a Factor Analysis or an Independent Component Analysis.

8. System for obtaining highly resolved flow field data representing the flow characteristics of fluid in a fluid system under analysis, the system comprising:

data capture means for capturing a set of flow predictor parameters for the fluid system under analysis from measured and/or estimated parameters of the fluid system under analysis,

data retrieval means for retrieving, from a reference database, highly resolved flow field data representing fluid flow simulation results of a set of reference fluid systems and flow predictor parameters of each of said set of reference fluid systems,

first calculation means for determining, by interpolation from said highly resolved flow field data and flow predictor parameters of said set of reference fluid systems, an interpolated set of flow predictor parameters which match, to within predetermined error margins, the said set of flow predictor parameters of the fluid system under analysis, and

second calculation means for determining, from said reference database, a set of highly resolved flow field data which corresponds to the said interpolated set of flow predictor parameters.

9. System according to claim 8, in which the first calculation means include means for performing a mathematical vector transform operation on the highly resolved flow field data and the flow predictor parameters in the database, and means for performing an interpolation on the mathematically transformed highly resolved flow field data.

10. System according to claim 9, in which the vector transform operation is a Principal Component Analysis, a Factor Analysis or an Independent Component Analysis.

11. Medical diagnostic device for analyzing fluid flow characteristics of biological fluid flowing in a biological conduit of a patient, the device comprising

diagnostic data capture means for acquiring parameters of the fluid flow through the biological conduit, and for generating therefrom a set of patient specific flow predictor parameters for the fluid flow through the biological conduit of the patient,

data retrieval means for retrieving high resolution flow field data and corresponding flow predictor parameters from a reference database,

interpolation means for, by interpolation of the high resolution flow field data and corresponding flow predictor parameters retrieved from the reference database, determining a set of interpolated flow predictor parameters which match, to within predetermined error margins, the said set of patient-specific flow predictor parameters,

reference transform data, which best fit said diagnostic transform data, and

second mathematical vector transform means for performing an inverse transform operation on said set of best-fit transform data, so as to produce a diagnostic set of high resolution flow field vector data corresponding to said diagnostic flow field vector data.