Paper
1 May 2002 Determination of in vivo photosensitizer concentrations using diffuse reflectance measurements and associative learning techniques
Author Affiliations +
Abstract
We examine the use of associative learning techniques as applied to diffuse reflectance spectra as possible methods for accurate in vivo drug concentration measurements. Aluminum phthalocyanine tetrasulphonate was added in known doses to sets of 8 homogenized tissue phantoms that covered a large range of optical properties. Measurements were made using a contact fiberoptic probe at several source/collection separations. Spectra were analyzed using partial least squares (PLS) and neural network (NN) regression using a leave 1/5-out-crossvalidation. Both the PLS and NN models accurately predicted the chromophore concentration within 15 percent. NN models provide lower prediction error, with optimal collection distance 3 mm form the source fiber. Predictions using the PLS models showed a systematic tissue-type dependency that correlated with optical pathlength of the reflected light. The advantages of this method are that it can be used for any chromophore, it requires only simple point spectroscopy and it is model- independent.
© (2002) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Robert A. Weersink, Brian C. Wilson, and Michael S. Patterson "Determination of in vivo photosensitizer concentrations using diffuse reflectance measurements and associative learning techniques", Proc. SPIE 4613, Optical Biopsy IV, (1 May 2002); https://doi.org/10.1117/12.465238
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Tissues

Tissue optics

Diffuse reflectance spectroscopy

Absorption

Picosecond phenomena

Chromophores

Optical properties

Back to Top