Intensity modulated near infrared spectroscopy: instrument design issues.

Tissue oxygenation instruments which rely on phase sensitive detection suffer from phase-amplitude crosstalk, i.e. the phase of the detected signal with respect to a reference signal is dependent on the average intensity of the light entering the photomultiplier tube (PMT). if an instrument that detects the phase of the scattered signal is to yield the phase accuracy required in order to provide useful clinical parameters, quantitative haemoglobin and oxy-haemoglobin concentrations, [Hb] and [HbO(2)] and mixed arterial-venous saturation (SavO2) all sources of phase-amplitude effects must be understood. The phase-amplitude effect has in the past been attributed to the fact that the rise time of the detector decreases with increasing light intensity. In this work an additional phase-amplitude effect in intensity modulated near infrared spectroscopy (IMNIRS) instrumentation is studied. The presence of a coherent interfering signal due to low level RF coupling at the detector output will corrupt the phase of the signal of interest and cause a phase-amplitude effect. Under certain conditions a relatively low level interfering RF signal (-90dBm) can introduce a significant (10%) error in the slope of the phase per unit distance plot. A comparison between measured and modelled phase distortion is presented and ways to reduce the effect discussed. In addition to phase-amplitude effects, the final accuracy of the quantitative measurements made by an IMNIRS instrument depends heavily on the calibration. Calibration of the measured phase and the AC and DC components of the detected Light must take into account distortions due to, (a) phase-amplitude crosstalk and system phase offset, (b) detector non-linearities, (c) variation in laser source intensity and phase with time and temperature, (d) optical probe light loss and (e) variations in detector sensitivity. Current instrument performance will. be presented and discussed.