Separation of indocyanine green boluses in the human brain and scalp based on time-resolved in-vivo fluorescence measurements.

Non-invasive detection of fluorescence from the optical tracer indocyanine green is feasible in the adult human brain when employing a time-domain technique with picosecond resolution. A fluorescence-based assessment may offer higher signal-to-noise ratio when compared to bolus tracking relying on changes in time-resolved diffuse reflectance. The essential challenge is to discriminate the fluorescence originating from the brain from contamination by extracerebral fluorescence and hence to reconstruct the bolus kinetics; however, a method to reliably perform the necessary separation is missing. We present a novel approach for the decomposition of the fluorescence contributions from the two tissue compartments. The corresponding sensitivity functions pertaining to the brain and to the extracerebral compartment are directly derived from the in-vivo measurement. This is achieved by assuming that during the initial and the late phase of bolus transit the fluorescence signal originates largely from one of the compartments. Solving the system of linear equations allows one to approximate time courses of a bolus for each compartment. We applied this method to repetitive measurements on two healthy subjects with an overall 34 boluses. A reconstruction of the bolus kinetics was possible in 62% of all cases.