Autofocusing in medical ultrasound: the scaled covariance matrix algorithm

• Published in: IEEE transactions on ultrasonics, ferroelectrics, and frequency control Vol. 50; no. 7; pp. 795 - 804
• Main Authors: Silverstein, S.D., Ceperley, D.P.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.07.2003; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Acoustic signal processing; Acoustics; Algorithm design and analysis; Algorithms; Analytical models; Biological and medical sciences; Computer Simulation; Covariance matrix; Error correction; Exact sciences and technology; Fundamental areas of phenomenology (including applications); Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Investigative techniques, diagnostic techniques (general aspects); Medical sciences; Miscellaneous. Technology; Physics; Quality Control; Regression Analysis; Reproducibility of Results; Scattering; Scattering, Radiation; Sensitivity and Specificity; Signal resolution; Speckle; Studies; Timing; Ultrasonic imaging; Ultrasonic investigative techniques; Ultrasonic transducers; Ultrasonography - methods; Human; Tissue; Error estimation; Focusing; Speckle; Signal processing; Medical imagery; Delay time; Covariance matrix; Acoustic image; Phase distortion
• ISSN: 0885-3010; 1525-8955
• DOI: 10.1109/TUFFC.2003.1214500

This work develops a class of ultrasound phase aberration correction/autofocusing algorithms that are based upon the properties of the covariance matrix of the channel signals for time-delay focused resolution/speckle cells. The scaled covariance matrix SCM algorithms are designed to blindly estimate and correct focusing timing errors due to thin layers of unanticipated fatty tissue located in the near field of the transducer array. An important aspect of the algorithm is that the scaling of the covariance matrix elements fundamentally establishes a channel independent phase reference relative to which the aberrant channel phases are estimated. The model development involved the combination of a rigorous mathematical analysis of the scattering of ultrasound in random scattering media and extensive statistical simulation studies with phase aberrations imposed upon both the transmit and received channel signals. Under the assumption of a near field aberration model, the statistical simulation analyses showed that the SCM algorithms in simulation are capable of accurately estimating relative time delay channel errors with RMS timing errors up to /spl sim/62 ns, with interchannel correlation lengths as short as 1.4 mm.