IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, zyxwvutsrqpon VOL. 43, NO. 8, AUGUST 1996 839 Dynamics of Temperature Dependent Optical Properties of Tissue: Dependence on Thermally Induced Alteration Ramtin Agah, A. H. Gandjbakhche, Massoud Motamedi, Ralph Nossal,” and R. F. Bonner zyx Abstruct- Thermal damage in heated bovine myocardial tis- sue is assessed from measured changes in total reflection and transmission of light. Mathematical expressions, based on random walk analysis of light propagation within tissue slabs, are used to relate the diffuse reflection and transmittance to the absorption coefficient, zyxwvutsrqp pa, and effective scattering coefficient, zyxwvutsr p; for samples of myocardial tissue which were subjected to rapid step changes in temperature. Time-dependent changes in p; indicate two processes, one with a fast and temperature-dependent rate the other with a slow and apparently temperature-independent rate. For final temperatures above 5623°C and for the first 500 zyxwvut s after the temperature change, the optical parameters are well fit by exponential forms that exhibit temperature-dependent time constants as predicted by Arrhenius reaction rate theory of thermal damage. The scattering changes are associated with an apparent activation energy, zyxwvutsrqpo AE, of 162 kJ/mole and a frequency constant, A, of 3x s-’. This method provides a means for estimating optical coefficients which are needed to assess laser tissue dosimetry. I. INTRODUCTION ANY therapeutic applications of lasers in medicine rely M on the photo-thermal effects induced by absorption of light in tissue. Two major effects are important in therapy, namely, photocoagulation and photothermal ablation. In both cases the optical properties of tissue play a dominant role. Tissue scattering strongly influences the distribution of light within tissues, while the local conversion of photon energy to thermal energy depends on tissue absorption. Because coagulation induces changes in optical properties, it is important to develop models which allow examination of how the optical coefficients change when tissue is heated, as during laser therapy [l]. Another reason to measure temporal changes in optical coefficients is to characterize the rate processes of coagulation. Such rates can be used in models of tissue thermal damage. Previously, morphological changes Manuscript received June 7, 1994; revised March 22, 1996. Asterisk R. Agah is with Baylor College of Medicine, Houston, TX 77030 USA. A. H. Gandjbakhche is with the Physical Sciences Laboratory, Division of Computer Research, National Institutes of Health, Bethesda, MD 20892 USA. M. Motamedi is with the Laser and Spectroscopy Program, University of Texas Medical Branch, Galveston, TX 77555 USA. *R. Nossal is with the Physical Sciences Laboratory, Division of Computer Research, National Institutes of Health, Bldg. 12A, zyxwvutsr rm. 2007, Bethesda, MD 20892 USA (e-mail: rjn@cu.nih.gov). R. F. Bonner is with the Biomedical Instrumentation and Engineering Program, National Center for Research Resources, National Institutes of Health, Bethesda, MD 20892 USA. indicates corresponding author. Publisher Item Identifier S 0018-9294(96)05567-X. observed by microscopy were used to determine the extent of heat damage to tissue samples and estimate coefficients of ther- mal damage processes [2]-[6]. This imethod suffers from a lack of quantifiable markers, and the subjective characterization of thermal damage diminishes the ability to obtain reproducible data from the same sample 171, [81. This paper describes a novel approach to quantitating ther- mal damage in tissue. The methodology involves relating real-time changes in total reflection and transmission of light to thermal lesions induced zyxw in vitro (in bovine myocardium) by a rapid, step increase in temperature. The sample preparation and experimental apparatus are discussed in Section 11. In Section 111-A the theoretical methods used to obtain optical coefficients from measured transmittance and reflectance are introduced. These involve an inverse solution of mathematical expressions derived from a photon random walk treatment of light diffusion within a slab of finite thickness [9]. A tissue thermal damage model based on Arrhenius reaction rate theory is presented in Section 111-B. Results are presented in Section IV, where the time de- pendent changes in absorption coefficient (pa) and effective scattering coefficient zyxwv (,U:) are shown. The changes in ,U: are used as a measure of protein denaturation and tissue coagulation, and the temperature-dependent rate processes characterizing the irreversible changes in scattering are used to determine the coefficients of the Arrhenius expression representing thermal damage. Finally, the advantages and limitations of this methodology arid the significance of the results are discussed in Section V. 11. EXPERIMENTAL METHODS Freshly excised bovine myocardium was immersed in 0.9% buffered saline and placed on ice immediately after harvest- ing. Within several hours, the samples were cleaned and a microtome was used to slice the tissue into thin square slabs of approximately 2 x 2 cm, with thicknesses ranging from 0.9 to 2.1 mm. The sliced specimens were then soaked in 0.9% buffered saline for 6-8 hours at 4OC to extract any residual blood, principally to minimize variability in measurements which might arise from differing tissue blood content. Each sample was then placed in a plastic tissue holder made from a tissue histology casse1te (“Tissue Path cassette,” Curtin Matheson Scientific, Inc.) and sandwiched between glass cover-slips of 75-pm thickness, the edges of the tissue 0018-9294/96$05.00 zyxwvut 0 1996 IEEE