Evaluation of a tissue-mimicking thermochromic phantom for radiofrequency ablation

Authors

  • Mikhail Andrew S.,

    1. Center for Interventional Oncology, Radiology and Imaging Sciences, Clinical Center, National Institutes of Health, Bethesda, Maryland 20892
    Search for more papers by this author
    • a)

      Author to whom correspondence should be addressed. Electronic mail: Andrew.mikhail@nih.gov; Telephone: +1-301-435-8945.

  • Negussie Ayele H.,

    1. Center for Interventional Oncology, Radiology and Imaging Sciences, Clinical Center, National Institutes of Health, Bethesda, Maryland 20892
    Search for more papers by this author
  • Graham Cole,

    1. Center for Interventional Oncology, Radiology and Imaging Sciences, Clinical Center, National Institutes of Health, Bethesda, Maryland 20892
    Search for more papers by this author
  • Mathew Manoj,

    1. Center for Interventional Oncology, Radiology and Imaging Sciences, Clinical Center, National Institutes of Health, Bethesda, Maryland 20892
    Search for more papers by this author
  • Wood Bradford J.,

    1. Center for Interventional Oncology, Radiology and Imaging Sciences, Clinical Center, National Institutes of Health, Bethesda, Maryland 20892
    Search for more papers by this author
  • Partanen Ari

    1. Center for Interventional Oncology, Radiology and Imaging Sciences, Clinical Center, National Institutes of Health, Bethesda, Maryland 20892 and Clinical Science MR Therapy, Philips, Andover, Massachusetts 01810
    Search for more papers by this author

Abstract

Purpose:

This work describes the characterization and evaluation of a tissue-mimicking thermochromic phantom (TMTCP) for direct visualization and quantitative determination of temperatures during radiofrequency ablation (RFA).

Methods:

TMTCP material was prepared using polyacrylamide gel and thermochromic ink that permanently changes color from white to magenta when heated. Color vs temperature calibration was generated in matlab by extracting RGB color values from digital photographs of phantom standards heated in a water bath at 25–75 °C. RGB and temperature values were plotted prior to curve fitting in mathematica using logistic functions of form f(t) = a + b/(1 + e(c(td))), where a, b, c, and d are coefficients and t denotes temperature. To quantify temperatures based on TMTCP color, phantom samples were heated to temperatures blinded to the investigators, and two methods were evaluated: (1) visual comparison of sample color to the calibration series and (2) in silico analysis using the inverse of the logistic functions to convert sample photograph RGB values to absolute temperatures. For evaluation of TMTCP performance with RFA, temperatures in phantom samples and in a bovine liver were measured radially from an RF electrode during heating using fiber-optic temperature probes. Heating and cooling rates as well as the area under the temperature vs time curves were compared. Finally, temperature isotherms were generated computationally based on color change in bisected phantoms following RFA and compared to temperature probe measurements.

Results:

TMTCP heating resulted in incremental, permanent color changes between 40 and 64 °C. Visual and computational temperature estimation methods were accurate to within 1.4 and 1.9 °C between 48 and 67 °C, respectively. Temperature estimates were most accurate between 52 and 62 °C, resulting in differences from actual temperatures of 0.6 and 1.6 °C for visual and computational methods, respectively. Temperature measurements during RFA using fiber-optic probes matched closely with maximum temperatures predicted by color changes in the TMTCP. Heating rate and cooling rate, as well as the area under the temperature vs time curve were similar for TMTCP and ex vivo liver.

Conclusions:

The TMTCP formulated for use with RFA can be used to provide quantitative temperature information in mild hyperthermic (40–45 °C), subablative (45–50 °C), and ablative (>50 °C) temperature ranges. Accurate visual or computational estimates of absolute temperatures and ablation zone geometry can be made with high spatial resolution based on TMTCP color. As such, the TMTCP can be used to assess RFA heating characteristics in a controlled, predictable environment.

Ancillary