Diffuse optical imaging
Visible and near infrared (NIR) light interact with biological tissue predominantly by absorption and elastic scattering. There are several physiologically interesting molecules which have characteristic absorption spectra at these wavelengths. In particular, the spectra of oxy-haemoglobin (HbO) and deoxy-haemoglobin (HHb) differ markedly, as do the oxygenation-dependent spectra of cytochrome oxidase. Haemoglobin provides an indicator of blood volume and oxygenation, whereas the cytochrome enzymes indicate tissue oxygenation.
Unfortunately, while most of the physiological information is contained in the absorption coefficient (the number of absorption events per unit length, µa), the scatter coefficient (the number of scattering events per unit length, µs) in tissue is generally considerably larger, so that signals measured over distances of a few millimetres or larger are dominated by diffuse light. The physics of light transport in tissue has been explained in a number of recent review articles, for example Boas et al. (2001a), Schweiger et al. (2003) and Dunsby and French (2003), and will not be covered in detail here.
The different absorption spectra of HbO and HHb are routinely exploited in physiological monitoring techniques such as pulse oximetry and near infrared spectroscopy (NIRS). Diffuse optical imaging techniques aim to process this information further and produce spatially resolved images. These images may display the specific absorption and scattering properties of the tissue, or physiological parameters such as blood volume and oxygenation. Diffuse optical imaging generally falls into one of two categories: topography or tomography. The distinction between these two techniques has become somewhat blurred, but we use the term optical topography when referring to methods which produce two-dimensional (2D) images of a plane parallel to the sources and detectors with limited depth information, and use optical tomography to describe techniques which generate full three-dimensional (3D) images from measurements taken from sources and detectors widely spaced over the surface of the object.
Progress since 1997
In 1997, we published two reviews of “optical imaging in medicine” in Physics in Medicine and Biology (Hebden et al. 1997, Arridge and Hebden 1997) which summarised the state-of-the-art of the field of medical optical imaging at that time. The first review (Hebden et al. 1997) examined instrumentation and experimental techniques, while the second review discussed modelling light transport in tissues and theoretical approaches to image reconstruction (Arridge and Hebden 1997). The focus of much of the experimental research at that time was on measuring and identifying minimally scattered photons, which (it was correctly concluded) cannot be applied to tissues more than a few millimetres thick. Improvement in spatial resolution was identified as a concern for diffuse optical imaging, and the potential for functional (and rapid) imaging of tissues (and of the brain in particular) was just beginning to be realised.
In this article, we examine the progress which has been made in diffuse optical imaging since the publication of the 1997 reviews. Recent advances in the field have largely focussed on the transfer of techniques from the laboratory to the clinic, and have led to the development of a broad variety of diagnostic applications, in particular imaging of the female breast and the adult and infant brain. While impressive, the progress has been evolutionary rather than revolutionary, although certain technological advances, in particular of laser diodes and fast desktop PCs, have enabled progress to have been more rapid than would otherwise have been the case.