Fluorescence Lifetime Imaging and Spectroscopy in Random Media

* Final gross prices may vary according to local VAT.

Get Access


With the development of near-infrared (NIR) laser diodes, the synthesis of fluorescent dyes with excitation and emission spectra in the NIR wavelength regime has accelerated in the past decade for microscopy applications. Owing to the window of low absorbance in tissues in this wave-length regime, an opportunity also exists for the deployment of fluorescent dyes as in vivo diagnostic agents. Figure 1.1 illustrates the typical absorbance spectra of tissues showing that at wavelengths less than 650 nm, hemoglobin absorbance provides the predominant attenuation of light in tissues, and above 900 nm water absorbance provides the predominant attenuation of light in tissues. In the “therapeutic window” of 650–900 nm, a window of low absorption exists in which light will be preferentially scattered over being absorbed. Consequently, it is possible to transmit multiply scattered NIR light across several centimeters of tissue. In addition, it is possible to excite NIR fluorophores deep within tissues and to collect the fluorescence re-emitted from the air-tissue interface. Since fluorescence provides a sensitive means for assessing local biochemistry via changes in quantum efficiency and lifetime, the ability to diagnose tissues based on spectroscopic evaluation of lifetime and quantum efficiency of exogenous diagnostic agents is possible. Agents whose emission characteristics vary with tissue pH [1, 2] and p02 [3] have been employed to detect diseased tissues by the nature of differing fluorescent properties as well as to provide diagnostic information regarding the diseased tissue volume. While typical contrast agents employed for the detection of diseased tissues depend on and are limited by the poor preferential uptake, the alteration of fluorescent properties provides a unique mechanism for inducing additional contrast [4]. For example, using time-gating to collect the long-lived fluorescence from hematoporphyrin derivative (HpD), Cubbeddu and co-workers [5, 6] were able to differentiate normal tissues from intradermally or subcutaneously implanted murine tumors in mice. More recently, it has been reported that the fluorescent decay of HpD is appreciably slower in experimental mice tumors than in their surrounding normal healthy tissues. Consequently, the use of a fluorescent dye may provide contrast owing to changes in fluorescent properties within tissue compartments. The difficulty, however, lies in understanding the use of multiply scattered excitation light to excite a fluorophore in the tissue, and secondly, to extract information of lifetime and quantum efficiency from the multiply scattered fluorescence detected at the air-tissue interface. If the optical properties of the tissue are spatially uniform and the fluorescent dye has constant fluorescent properties, then the problem is one of fluorescent lifetime spectroscopy in tissues. However, if the detection of diseased tissues is to be tackled, the problem becomes one of fluorescent lifetime spectroscopic imaging, since optical properties and fluorescent properties of diagnostic agents may vary with spatial location. In this chapter, frequency domain photon migration fluorescence imaging is described as a method for generating an optical map or image of fluorescent lifetime and quantum efficiency from exterior measurements of modulation phase and modulation amplitude on tissues or highly scattering media. In Sect. 1, the theory behind the propagation of excitation light and generation of fluorescent light within scattering media such as tissues is presented. Section 2 describes experimental measurements which show that the delay in phase and decrease in amplitude of fluorescence measured in simulated tissue phantoms varies as a function of dye lifetime. Section 3 describes the general theory behind the derivation of an optical property map from measurements of modulation phase and amplitude of fluorescent light detected at the tissue surface, and Sect. 4 presents actual images generated from simulated measurements of modulation phase and amplitude. Finally, the prognosis for moving these theoretical and computational studies into an experimental demonstration is commented upon.