U.S. flag An official website of the United States government

U.S. Food and Drug Administration

Photoacoustic Imaging Phantoms for Assessing Image Quality and Oximetry Performance

Catalog of Regulatory Science Tools to Help Assess New Medical Devices 

 

This regulatory science tool presents a set of tissue-mimicking phantoms suitable for benchtop performance assessment of photoacoustic imaging (PAI) devices. 

 

Technical Description

The tissue-mimicking phantoms include: 

  1. A tissue-mimicking material (TMM) with tunable, biologically relevant optical and acoustic properties that can be used to construct imaging phantoms, 
  2. a set of phantom-based test methods for evaluating photoacoustic image quality, and 
  3. a blood flow phantom test method for evaluating PAI oximetry measurement accuracy. 

The TMM is comprised of polyvinyl chloride plastisol (PVCP), which has higher mechanical strength and temporal stability than traditional hydrogel TMMs. PVCP formulations can be tuned to simulate properties of different tissues, including breast-like formulations. Image quality test methods use PVCP phantoms containing targets suitable for evaluating the following image quality characteristics: 

  • axial/lateral/elevational spatial resolution, 
  • uniformity, 
  • sensitivity, 
  • maximum imaging depth, 
  • spatial measurement accuracy, and 
  • ultrasound-PAI co-registration. 

The oximetry test method uses a PVCP phantom connected to a blood flow circuit, where blood oxygen saturation (SO2) can be tuned using a membrane oxygenator. PAI SO2 values are compared to readings from a gold-standard, clinical-grade CO-oximeter. 

Intended Purpose 

These phantoms are intended for objective, quantitative assessment of photoacoustic imaging (PAI) device performance in terms of image quality and oximetry measurement accuracy. These tools may be used to support PAI device development, design optimization, device intercomparison, benchmarking, calibration, quality assurance/control, and constancy testing. This includes evaluation of design consequences for system hardware (transducer or optical source parameters) and software (reconstruction, oximetry, and fluence correction algorithms).

While PAI devices are being developed for many potential clinical applications, breast cancer imaging is an important indication and is the first application for which a PAI device has been approved by the FDA (product code: QNK). Thus, specific phantom formulations that recapitulate properties of breast tissue have been developed. The phantom material can be readily tuned to mimic different tissue types to match an intended application, or to simulate properties of a generic soft tissue.

Testing

TMM optical and acoustic properties have been rigorously characterized and are consistent with reported literature values for breast tissue. Specific properties assessed include: 

  • optical absorption coefficient, 
  • optical reduced scattering coefficient, 
  • acoustic attenuation, and 
  • speed of sound. 

TMM repeatability was evaluated by characterizing eight disk-shaped samples. Samples were also used to assess TMM temporal stability by measuring changes in mass, optical properties, and acoustic properties over 12 weeks. After 12 weeks, samples showed mass loss <0.75%, and no monotonic trends in optical or acoustic properties were observed [1]. Further repeatability testing of the same samples over 2 years showed mass loss < 1.5%, optical property changes < 20%, acoustic attenuation changes < 10%, and speed of sound changes < 1% (unpublished data).

PVCP phantom utility was demonstrated by evaluating performance of a custom, modular PAI system similar to devices described in the literature [2]. Phantom testing compared image quality of the modular PAI system equipped with different array transducers to represent different device designs. Where appropriate, image quality characteristics were compared against values for interleaved B-mode ultrasound images, including 3D spatial resolution and spatial measurement accuracy. 

PVCP phantoms were used to evaluate oximetry measurement accuracy of a custom PAI system [3]. Photoacoustic SO2 measurements were compared against CO-oximeter SO2 readings and shown to be readily corrupted by spectral coloring artifacts. Some devices compensate for these effects using fluence correction algorithms, which can also be evaluated using this method [4]. 

Limitations

While the phantom geometry is easily adapted or generalized for various imaging system configurations, specific phantom designs/embodiments described in published articles are best suited to linear-array transducer geometries commonly used in handheld imaging probes. TMM optical properties have only been characterized over 400-1100 nm, while acoustic properties have only been characterized over 4-9 MHz. When testing PAI devices with operating parameters outside those ranges, phantom properties should be characterized and compared to reported tissue values to determine suitability for a given application. PVCP phantoms may experience slight plasticizer exudation over time and should be contained in housings using compatible protective membranes (“acoustic windows”). PVCP phantoms should generally be handled with latex or nitrile gloves and according to safety data sheets for TMM components.

Supporting Documentation

  1. Vogt, W. C., Jia, C., Wear, K. A., Garra, B. S., & Joshua Pfefer, T. (2016). Biologically relevant photoacoustic imaging phantoms with tunable optical and acoustic properties. Journal of biomedical optics, 21(10), 101405. https://doi.org/10.1117/1.JBO.21.10.101405
  2. Vogt, W. C., Jia, C., Wear, K. A., Garra, B. S., & Pfefer, T. J. (2017). Phantom-based image quality test methods for photoacoustic imaging systems. Journal of biomedical optics, 22(9), 1–14. https://doi.org/10.1117/1.JBO.22.9.095002
  3. Vogt, W. C., Zhou, X., Andriani, R., Wear, K. A., Pfefer, T. J., & Garra, B. S. (2019). Photoacoustic oximetry imaging performance evaluation using dynamic blood flow phantoms with tunable oxygen saturation. Biomedical optics express, 10(2), 449–464. https://doi.org/10.1364/BOE.10.000449
  4. Zhou, X., Akhlaghi, N., Wear, K. A., Garra, B. S., Pfefer, T. J., & Vogt, W. C. (2020). Evaluation of Fluence Correction Algorithms in Multispectral Photoacoustic Imaging. Photoacoustics, 19, 100181. https://doi.org/10.1016/j.pacs.2020.100181
  5. Patent: PVCP Phantoms and Their Use, US 9,920,188-B2  

Contact

Tool Reference

  • In addition to citing relevant publications please reference the use of this tool using RST24MD02.01