Arjun G. Yodh

Fluorescent heterogeneities in turbid media: limits for detection, characterization, and comparison with absorption

The fundamental limits for detection and characterization of fluorescent (phosphorescent) inhomogeneities embedded in tissuelike highly scattering turbid media are investigated. The absorption and fluorescence contrast introduced by exogenous fluorophores are also compared. Both analyses are based on practical signal-to-noise ratio considerations. For an object with fivefold fluorophore concentration and lifetime contrast with respect to the background tissue, we find the smallest detectable fluorescent object at 3-cm depth in tissuelike turbid media to be ~0.25 cm in radius, whereas the smallest characterizable object size is ~0.75 cm in radius, given a model with 1% amplitude and 0.5 degrees phase noise. We also find that, for fluorescence extinction coefficients epsilon </= 0.5 x 10(5) cm(-1) M(-1), the fluorescence measurement mode is superior to the absorption mode for detecting an inhomogeneity. The optimal choice of modulation frequency for the frequency-domain fluorescence measurements is also discussed.

Optimal selection of frequencies for diffuse optical tomography

The use of diffuse photon density waves (DPDW) at multiple modulation frequencies is a significant tool for Diffuse Optical Tomography (DOT). Use of higher frequencies offers higher contrast between scattering and absorbing objects and allows higher resolution to be achieved. Here we focus on the optimal use of frequencies in simultaneously resolving absorbing and scattering objects using simulated data.

Optimization of the Source-detector Geometry for Diffuse Optical Tomography

We use simulation and singular value analysis to optimize source-detector geometry for breast imaging in the parallel plane geometry.

Performance of the linear perturbation tomography with highly heterogeneous media under the PI approximation

The theoretical and instrumental advances accomplished the last decade in the field of imaging using NIR light has lately led to the development of imaging devices aiming in investigating the clinical capacity of the technique1,2,3.

Near infrared imaging of brains

The authors describe a near-infrared imaging instrument capable of providing spectroscopic information over a spatially localized region of the brain. This regional imager is a frequency domain, multiwavelength (779 and 834 nm) system with multiple source (12) and detector (4) positions. The modulation frequency is selectable, ranging from 50 to over 400 MHz. The regional imager offers the possibility for reconstruction of optical properties with linear resolutions of /spl sim/1 cm, beneath an area of /spl sim/40 cm/sup 2/ in size, within a short period of time (3 min). Clinical applications include evaluation of critical care conditions such as hydrocephalus and hematoma, as well as monitoring blood oxygenation during heart surgery and cognition studies. For real-time analysis of a measurement, the authors use a backprojection algorithm to reconstruct images. In the authors backprojection approach, they derive an absorption coefficient /spl mu//sub s/ and scattering coefficient /spl mu//sub s/ for each source-detector pair by numerically solving the semi-infinite solution for the values of /spl mu//sub s/ and /spl mu//sub s/ using the measurements from the source-detector pair. These values are then back-projected using weights that are essentially the three-point Greens function. The authors are currently investigating use of the Rytov approximation for image reconstruction in a reflectance geometry.