Deformation models and correlation analysis in elastography.

M Bilgen; M F Insana

doi:10.1121/1.414865

Home/Archive/PubMed

PubMed · 8642127

Deformation models and correlation analysis in elastography.

M Bilgen·M F Insana

View original source ↗Export metadata ↓

Abstract

Cross-correlation functions are derived with the purpose of determining how strain inhomogeneities affect the displacement estimates used in ultrasound-based elastography. Variations in the strain profile occur in most imaging situations and are caused by fluctuations in the stress field or elastic modulus of the sample. An analytical framework for developing signal processing strategies in elastography is described, and the limitations of correlation-based methods for measuring displacements in tissuelike media caused by static compression are emphasized. This paper includes (1) an accurate approximation for an inverse coordinate transformation that release pre- and postcompression reflectivity profiles of the media, (2) a derivation of the echo-signal cross-correlation function in media with deterministic or stochastic strain profiles; (3) mathematical and graphical descriptions of the consequences that nonuniformities in the strain profile impose upon the uncertainty of displacement estimation; and (4) a demonstration of the advantages of echo signal conditioning and ultrasonic-pulse shaping to reduce the nonstationary effects that attenuate the cross-correlation peak and reduce the signal-to-noise ration for displacement estimation.

Explore related subjects

Elasticity Humans Models, Theoretical Comparative Study Journal Article Research Support, Non-U.S. Gov't Research Support, U.S. Gov't, P.H.S.

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

M Bilgen, M F Insana. 1996. Deformation models and correlation analysis in elastography.. https://doi.org/10.1121/1.414865

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed1999

Molecular theory of lipid-protein interaction and the Lalpha-HII transition.

We present a molecular-level theory for lipid-protein interaction and apply it to the study of lipid-mediated interactions between proteins and the protein-induced transition from the planar bilayer (Lalpha) to the inverse-hexagonal (HII) phase. The proteins are treated as rigid, membrane-spanning, hydrophobic inclusions of different size and shape, e.g., "cylinder-like," "barrel-like," or "vase-like." We assume strong hydrophobic coupling between the protein and its neighbor lipids. This means that, if necessary, the flexible lipid chains surrounding the protein will stretch, compress, and/or tilt to bridge the hydrophobic thickness mismatch between the protein and the unperturbed bilayer. The system free energy is expressed as an integral over local molecular contributions, the latter accounting for interheadgroup repulsion, hydrocarbon-water surface energy, and chain stretching-tilting effects. We show that the molecular interaction constants are intimately related to familiar elastic (continuum) characteristics of the membrane, such as the bending rigidity and spontaneous curvature, as well as to the less familiar tilt modulus. The equilibrium configuration of the membrane is determined by minimizing the free energy functional, subject to boundary conditions dictated by the size, shape, and spatial distribution of inclusions. A similar procedure is used to calculate the free energy and structure of peptide-free and peptide-rich hexagonal phases. Two degrees of freedom are involved in the variational minimization procedure: the local length and local tilt angle of the lipid chains. The inclusion of chain tilt is particularly important for studying noncylindrical (for instance, barrel-like) inclusions and analyzing the structure of the HII lipid phase; e.g., we find that chain tilt relaxation implies strong faceting of the lipid monolayers in the hexagonal phase. Consistent with experiment, we find that only short peptides (large negative mismatch) can induce the Lalpha --> HII transition. At the transition, a peptide-poor Lalpha phase coexists with a peptide-rich HII phase.

Elasticity↗

PubMed1998

Multicycle mechanical performance of titanium and stainless steel transpedicular spine implants.

STUDY DESIGN: This was a prospective in vitro study comparing titanium alloy and stainless steel alloy in transpedicular spine implants from two different manufactures. OBJECTIVE: To compare the multicycle mechanical performance of these two alloys, used in each of two different implant designs. SUMMARY OF BACKGROUND DATA: Transpedicular spine implants primarily have been manufactured from stainless steel, but titanium alloy offers imaging advantages. However, the notch sensitivity of titanium alloy has caused concern regarding how implants made from this material will compare in stiffness and fatigue life with implants made from stainless steel. METHODS: Twenty-four implants (two alloys, two designs, six implants per group) were mounted in machined polyethylene wafers and repetitively loaded (up to 1 million cycles) from 80 N to 800 N using a 5-Hertz sinusoidal waveform. Load and displacement data were automatically and periodically sampled throughout the entire test. RESULTS: Implant stiffness increased with cycle load number, reached a steady state, then declined just before fatigue failure. Stiffness varied less in titanium transpedicular spine implants than in their stainless counterparts. All stainless steel implant types were stiffer (steady-state value, P < 0.0001) than their titanium alloy counterparts. One titanium implant design failed with fewer (P < 0.05) load cycles than its stainless steel counterpart, whereas a stainless steel implant of another design failed with fewer (P < 0.002) load cycles than its titanium counterpart. Overall, fatigue life, i.e., the total number of load cycles until failure, was related to implant type (P < 0.0001), but not to implant material. CONCLUSIONS: A transpedicular spine implant's fatigue lifetime depends on both the design and the material and cannot be judged on material alone. Stainless steel implants are stiffer than titanium alloy implants of equal design and size; however, for those designs in which the fatigue life of the titanium alloy version is superior, enlargement of the implant's components can compensate for titanium's lower modulus of elasticity and result in an implant equally stiff as its stainless steel counterpart. Such an implant made from titanium alloy would then be clinically preferable because of titanium's previously reported imaging advantages.

Elasticity↗

PubMed1998

Acyl chain-length asymmetry alters the interfacial elastic interactions of phosphatidylcholines.

Phosphatidylcholines (PCs) with stearoyl (18:0) sn-1 chains and variable-length, saturated sn-2 acyl chains were synthesized and investigated using a Langmuir-type film balance. Surface pressure was monitored as a function of lipid molecular area at various constant temperatures between 10 degrees C and 30 degrees C. Over this temperature range, 18:0-10:0 PC displayed only liquid-expanded behavior. In contrast, di-14:0 PC displayed liquid-expanded behavior at 24 degrees C and 30 degrees C, but two-dimensional phase transitions were evident at 20 degrees C, 15 degrees C, and 10 degrees C. The average molecular area of 18:0-10:0 PC was larger than that of liquid-expanded di-14:0 PC at equivalent surface pressures, and the shapes of their liquid expanded isotherms were somewhat dissimilar. Analysis of the elastic moduli of area compressibility (Cs(-1)) as a function of molecular area revealed shallower slopes in the semilog plots of 18:0-10:0 PC compared to di-14:0 PC. At membrane-like surface pressures (e.g., 30 mN/m), 18:0-10:0 PC was 20-25% more elastic (in an in-plane sense) than di-14:0 PC. Other PCs with varying degrees of chain-length asymmetry (18:0-8:0 PC, 18:0-12:0 PC, 18:0-14:0 PC, 18:0-16:0 PC) were also investigated to determine whether the higher in-plane elasticity of fluid-phase 18:0-10:0 PC is a common feature of PCs with asymmetrical chain lengths. Two-dimensional phase transitions in 18:0-14:0 PC and 18:0-16:0 PC prevented meaningful comparison with other fluid-phase PCs at 30 mN/m. However, the Cs(-1) values for fluid-phase 18:0-8:0 PC and 18:0-12:0 PC were similar to that of 18:0-10:0 PC (85-90 mN/m). These values showed chain-length asymmetrical PCs to have 20-25% greater in-plane elasticity than fluid-phase PCs with mono- or diunsaturated acyl chains.

Elasticity↗