Kalman filtering for flash artifact suppression in ultrasound imaging
AssigneeSiemens Medical Solutions USA, Inc.
InventorsLei Sui, Kathryn Ozgun
Life of this patent
Estimating the term…
Abstract
Kalman filtering, including a model of system dynamics, is used to identify flash artifact. The Kalman filtering predicts the velocity or powers based on a past sequence of velocity or power. The prediction defines a confidence interval. Any current measures surpassing the confidence interval are identified as flash artifact and suppressed.
Claims20
A method for flash artifact suppression by an ultrasound scanner, the method comprising:
- scanning, by the ultrasound scanner, a patient at different times;
- estimating first velocity or power values from scan data from the scanning of the patient, the velocity or power values representing motion in the patient at the different times;
- predicting second velocity or power values based on the first velocity or power values from prior ones of the different times and based on a model of system dynamics;
- determining statistical confidence intervals from the second velocity or power values;
- replacing the first velocity or power values for current ones of the different times when the first velocity or power values exceed the statistical confidence intervals, the first velocity or power values replaced with replacement velocity or power values;
- and
- generating a flow image, on a display, from the replacement velocity or power values.
A method of flash artifact suppression for ultrasound flow imaging, the method comprising:
- detecting the flash artifact from a predicted flow value of a Kalman filter;
- suppressing the flash artifact;
- and
- generating a flow image on a display, the flow image based on the suppression of the flash artifact.
An ultrasound system for artifact suppression in ultrasound imaging, the ultrasound system comprising:
- a transducer and beamformer for scanning a scan region;
- a Doppler estimator configured to estimate, from the scanning, first estimates in the scan region over time;
- a Kalman filter configured to predict second estimates from the first estimates;
- an image processor configured to detect the artifact in the first estimates from the second estimates and to suppress the artifact in the first estimates;
- and
- a display configured to display an image from the first estimates with the artifact suppressed.
Dependencies are inferred from the claim text.