Imaging Performance of Quantitative Transmission Ultrasound

M. Lenox, James Wiskin, Matthew A. Lewis, Stephen Darrouzet, D.t. Borup, Scott S. HsiehInternational Journal of Biomedical Imaging 2015 · DOI

TL;DR

This paper introduces and evaluates a Quantitative Transmission Ultrasound (QTUS) system, a novel tomographic imaging modality capable of generating 3D maps of sound speed and reflection properties. The system demonstrates high accuracy in speed-of-sound measurements and promising spatial resolution, paving the way for improved detection of pathologies like calcifications.

What Problem Does This Paper Address?

Traditional ultrasound imaging relies heavily on reflection, which can be limited in its ability to characterize tissue properties beyond simple impedance mismatches. While transmission ultrasound has been explored for some time, recent advancements in hardware and algorithms have opened doors for more sophisticated applications. The challenge lies in developing and validating new tomographic ultrasound techniques like QTUS, which combine both transmission and reflection data. Specifically, this paper addresses the need for standardized methods to evaluate the performance of such systems, focusing on their accuracy in measuring sound speed, spatial resolution, and contrast detection, particularly for applications like identifying calcifications in breast lesions.

Methods

The QTUS system employs a U-shaped arrangement of a plane wave transmitter and a high-resolution receiver. Transmission data is acquired by rotating the transmitter-receiver pair around the subject, generating multiple projection views. This data is then processed using 3D nonlinear inverse scattering algorithms to reconstruct quantitative maps of sound speed (at 1.5 MHz) and attenuation.

For reflection imaging, three transducers with different focal lengths are used. These are fired between transmission measurements, and the resulting reflection data is compounded and corrected for refraction using the speed map from the transmission phase. This creates a high-resolution, spatially coregistered reflection image.

To evaluate performance, the authors developed several phantoms and measurement protocols:

  • Speed Phantom: Used to assess quantitative accuracy of speed-of-sound measurements.
  • Spatial Resolution Phantoms: Employed a nautilus spiral of monofilament for transmission and spherical beads for reflection to measure spatial resolution and accuracy. A novel "square step response" method was proposed to overcome limitations of previous techniques.
  • Slice Thickness Phantom: A dual inverted comb phantom was used to determine slice thickness in both modes.
  • Uniformity Phantom: A uniform polyurethane block was used to assess speed uniformity.
  • Contrast Phantom: A phantom with styrene rods was used to evaluate contrast resolution and contrast-to-noise ratio (CNR).

Datasets

The paper does not discuss training datasets. Instead, it focuses on evaluating a prototype QTUS system using custom-designed phantoms. These phantoms serve as benchmarks for assessing various performance metrics, including:

  • Speed Phantom: Contained inclusions of known speeds to simulate soft tissue pathologies.
  • Spatial Resolution Phantoms: Utilized monofilament (608 microns) and glass beads (500 microns) to test resolution limits.
  • Contrast Phantom: Featured styrene rods of varying diameters (down to 400 microns) to assess contrast detection capabilities.

These phantoms represent a new benchmark for evaluating tomographic ultrasound systems, offering a more comprehensive assessment than previously available.

Key Insights and Ideas

  • Synergistic Imaging: The core innovation is the seamless integration of quantitative transmission (speed-of-sound) and high-resolution reflection imaging within a single system. This co-registration allows for a more complete characterization of anatomical structures and pathologies.
  • Accurate Speed Measurement: The system demonstrates remarkable accuracy in measuring sound speed, achieving within 1% of actual values for typical tissue-like materials. This is crucial for applications where subtle changes in tissue stiffness are indicative of disease.
  • Novel Resolution Evaluation: The proposed "square step response" method for spatial resolution measurement is a significant contribution, offering a more robust and less biased approach compared to prior techniques.
  • Sub-Millimeter Contrast Detection: The system shows excellent contrast resolution, particularly in reflection mode, capable of detecting structures down to 400 microns. This is highly relevant for identifying microcalcifications, a key indicator of early breast cancer.
  • Elimination of Speckle: The compounding and correction techniques used in reflection imaging effectively reduce speckle, leading to cleaner images that are more interpretable for anatomical reference.

Key Results

  • Speed-of-Sound Accuracy: The system accurately determined speed of sound within 1% of the actual value for materials simulating water and faster tissues.
  • Transmission Resolution: Intrinsic resolution in transmission mode averaged 2.335 mm in both X and Y directions. Slice thickness was measured at 2 mm.
  • Reflection Resolution: Average resolution was 543 microns in the X direction and 557 microns in the Y direction, with performance improving to 453 microns (X) and 482 microns (Y) in areas outside the exact center. Slice thickness in reflection mode was 6 mm.
  • Contrast Resolution: The system demonstrated contrast resolution better than 400 microns in reflection mode. In transmission mode, it could detect objects down to 0.4 mm with expected degradation due to partial volume effects.
  • Contrast-to-Noise Ratio (CNR): In speed mode, CNR remained above 10:1 even for objects as small as 400 microns. Reflection mode showed consistently high CNR across the tested range.
  • Uniformity: Speed uniformity was measured at 3.1%.

Limitations and Future Work

  • Phantom-Based Evaluation: The current evaluation relies on phantoms. While these are well-designed, clinical validation is the next crucial step.
  • Limited Clinical Data: The paper acknowledges that future work should include clinical trials to perform receiver operating characteristic (ROC) analysis for evaluating calcifications.
  • Further Phantom Studies: The authors suggest more phantom studies with smaller structures to fully define the technology's limits.
  • Reflection Uniformity: The paper notes that uniformity for reflection images is not well-defined and requires further investigation.
  • PSF Modeling: For the purpose of this paper, point-spread-function (PSF) modeling was disabled during reconstruction to accurately measure intrinsic resolution. Future work could explore how PSF modeling might further enhance clinical image quality.

Summary updated

Abstract

Quantitative Transmission Ultrasound (QTUS) is a tomographic transmission ultrasound modality that is capable of generating 3D speed-of-sound maps of objects in the field of view. It performs this measurement by propagating a plane wave through the medium from a transmitter on one side of a water tank to a high resolution receiver on the opposite side. This information is then used via inverse scattering to compute a speed map. In addition, the presence of reflection transducers allows the creation of a high resolution, spatially compounded reflection map that is natively coregistered to the speed map. A prototype QTUS system was evaluated for measurement and geometric accuracy as well as for the ability to correctly determine speed of sound.
BibTeX
@article{lenox2015imaging,
  author = {M. Lenox and James Wiskin and Matthew A. Lewis and Stephen Darrouzet and D.T. Borup and Scott S. Hsieh},
  title = {{Imaging Performance of Quantitative Transmission Ultrasound}},
  journal = {International Journal of Biomedical Imaging},
  year = {2015},
  doi = {10.1155/2015/454028},
}
Figure