Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by synovial inflammation. The current treatment paradigm for earlier, more aggressive therapy places importance on development of functional imaging modalities, capable of quantifying joint changes at the earliest stages. Diffuse optical tomography (DOT) has shown great promise in this regard, due to its cheap, non-invasive, non-ionizing and high contrast nature. Underlying pathological activity in afflicted joints leads to altered optical properties of the synovial region, with absorption and scattering increasing. Previous studies have used these optical changes as features for classifying diseased joints from healthy. Non-tomographic, single wavelength, continuous wave (CW) measurements of trans-illuminated joints have previously reported achieving this with specificity and sensitivity in the range 80 – 90% . A single wavelength, frequency domain DOT system, combined with machine learning techniques, has been shown to achieve sensitivity and specificity in the range of 93.8 – 100% . A CW system is presented here which collects data at 5 wavelengths, enabling reconstruction of pathophysiological parameters such as oxygenation and total hemoglobin, with the aim of identifying localized hypoxia and angiogenesis associated with inflammation in RA joints. These initial studies focus on establishing levels of variation in recovered parameters from images of healthy controls.