Quantitative Optical Sensing for Non-Invasive Clinical Characterization of Biological Tissues.

Abstract

It is well known that changes in tissue morphology and/or biochemistry can affect tissue function. Characterizing these changes in tissue function through non-invasive and label-free assessment can inform clinical practice and improve patient outcomes. In this thesis, we employ non-invasive, quantitative, label-free, portable, and clinically-compatible reflectance and fluorescence spectroscopic technology for use in two clinical challenges: (1) improved detection of pancreatic disease and (2) post-implantation monitoring of tissue-engineered construct wound healing in an in situ murine model.

(1) Only 6% of pancreatic cancer patients survive 5 years after diagnosis, making it the 4th leading cause of cancer death in the United States. To improve detection of pancreatic cancer, we studied the diagnostic utility of optical spectroscopy to detect pancreatic disease in 5 Stages, with Stages 1 and 2 previously reported. Stage 1 showed that ex vivo measurements of human adenocarcinoma tissue correspond well to in vivo measurements from a tumor xenograft in a murine model. Stage 2 showed that malignant tissues measured ex vivo distinguish malignant and benign tissues. In this thesis, we discuss Stages 3-5. In Stage 3, a photon-tissue interaction (PTI) model was verified with measurements from tissue-simulating phantoms and validated with measurements from a subset of ex vivo human tissues collected in Stage 2. We show that a calibrated PTI model consistently extracts biologically-relevant optical tissue scattering parameters in the presence of variable hemoglobin absorption. In Stage 4, we perform the first ever, to our knowledge, in vivo feasibility study employing optical steady-state spectroscopy to detect malignant tissues during open surgery. In Stage 5, we investigate time-resolved fluorescence spectroscopy ex and in vivo to improve pancreatic disease classification. Furthermore, we show the first ever human pancreatic tissue measurements with an endoscopically-compatible fiber-optic probe. (2) Regulatory approval for tissue-engineered combinational devices, including tissue constructs developed for human implantation, requires reliable methods to assess post-implantation wound healing in vivo, of which none currently exist. In this thesis, we investigate diffuse reflectance spectroscopy to detect hallmarks of graft wound healing, including tissue revascularization, cell proliferation, and cell density, based on construct absorption and scattering properties.