Enhancement of near-infrared diffuse optical tomography for prostate cancer imaging.

摘要

Scope and method of study. The objective of this study is to enhance the functional imaging modality of Diffuse Optical Tomography (DOT) for the early diagnosis and accurate detection of prostate cancer. Enhancement approaches such as spatial prior extracted from Trans-Rectal Ultrasound (TRUS), spectral prior pre-calibrated with the absorption chromophores in biological tissue and metabolic fluorescence emitting biomarker has been implemented by either instrumentation or computational modeling techniques.;Findings and conclusions. A hard prior based hierarchical reconstruction algorithm for a TRUS-DOT combined system is developed and validated by both simulation and tissue phantom experiments.;A Parametric Recovery Uncertainty Level (PRUL) model is derived for DOT reconstruction based on measurements at both single and multiple wavelengths. The model indicates that steady state DOT system could produce reconstruction results comparable to frequency domain systems and direct current measurement components are not redundant in frequency domain reconstruction. The model can also be used to optimize the wavelength combination in multispectral optical imaging systems.;A rapid DOT system is fabricated based on a wavelength-swept light source. The system is validated by phantom experiments and is readily extendable for fluorescence signal acquisition.;A geometric-differential-sensitivity reconstruction algorithm is developed for reflectance imaging geometry in DOT for accurate target depth recovery.;The experiment setup for FDOT validation is designed. Preliminary simulation results support the hypothesis that the reverse fluorohphore uptake can be recovered within our imaging geometry.