|
Medical diagnostic techniques
based on near infrared (NIR) transillumination were first
introduced more than 70 years ago to detect breast cancer.
Although NIR light penetrates tissue to depths of several
centimeters, early methods were not successful due to the fact
that these approaches were qualitative and did not account for
distortions from multiple light scattering.
Recent advances
in temporal- and spatial- frequency-domain “photon migration”
now make it possible to separate light absorption from
scattering in thick tissues. Temporal frequency-domain methods
measure the phase shift and amplitude of MHz - GHz
intensity-modulated waves, while spatial frequency-domain
techniques utilize structured light patterns to form wide-field
images of tissue optical properties. Both approaches are
based on comparing measured data with radiative transport models
to acquire spectra and form images, i.e. diffuse optical
spectroscopic imaging (DOSI).
This talk
reviews principles of light propagation in tissue and describes
the development of DOSI for non-invasively characterizing tissue
structure and biochemical composition. Particular emphasis is
placed on broadband methods for quantitatively recovering NIR
absorption and scattering spectra. These data are used to
determine the tissue concentration of deoxygenated hemoglobin,
oxygenated hemoglobin, methemoglobin, lipid, and water, as well
as the tissue “scatter power”. Clinical study results are shown
highlighting the sensitivity of broadband DOSI to metabolic
changes in breast cancer and in therapeutic drug monitoring.
Broadband spatial frequency-domain imaging is used in
pre-clinical animal models to dynamically map intrinsic brain
signals and monitor the efficacy of chemotherapeutic agents.
These findings will be placed in the context of conventional
imaging methods, such as MRI, in order to assess the current and
future role of diffuse optics in medical imaging. |