Near-Infrared Spectroscopy in Functional Activation Studies: New Approaches H . O b r i g 1, C . H i r t h 1, J . R u b e n 1, U . D i r n a g l I , A .Vi l l r i nge r 1. in cooperation with H. Wabnitz 2, D.Grosenick 2, H.Rinneberg 2. Dept. of Neurology, Charit6 Humboldt University 10098 Berlin F.R. Germany PhysikalischTechnische Bundesanstalt (PTB) 1058 7 Berlin F.R. Germans Background: Near-infrared spectroscopy (NIRS) is based On the transparency of skin, skull and brain tissue to light between 700-1100 nm. The attenuation of the light depends on two parameters: absorption by chromophores and scattering by the tissue. Current in vivo NIRS approaches assess changes in chromophore concentration at discrete wavelengths. With a NIRO-500 system (Hamamatsu, Japan) we recently demonstrated an increase in [oxy-Hb] and a decrease in [deoxy-Hb] during motor stimulation, when measuring over the primary motor cortex (Obrig et al. 1995). Purpose: Two major problems are associated with the current NIRS approach: (1) It is not possible to assess changes in scattering properties of the brain tissue. (2) Spectral resolution is low (4 wavelengths). Recently two methods have been proposed to overcome these shortcomings. The purpose of the present study was to implement those techniques in our experimental set-up and to assess their sensitivity to changes in response to functional brain activation. Methods: For both experiments the optical probes were positioned over the left hemisphere in a C3' position, cor-responding to the hand region of the primary motor cortex according to 10-20 system. Interoptode distance was 3.5 cm. The stimulation paradigm consisted of a 20s unilateral finger opposition task ipsiand contralateral to the optodes fol-lowed by a 40 s period of rest. 10-15 cycles of either condition were performed in each subject. (1) Time correlated ~ h o t o n counting technique (TCSPC): time resolution: 35ps. n= 4 ss. Light source: Ti mode-locked Sapphire laser tuned to 775, 805 and 835 nm successively. Determination of the mean time of flight and time-integrated (relative) transmittance from distributions of times of flight measured at a repetition time of 1 s. (2) Whole spectrtma approach: n= 9 ss. Lightsource: Halogen bulb, with a spectral range from 600-1100nm. Assessment of the whole spectrum of the reflected light by means of a spectrograph and a CCD Camera with a sampling rate of I s. Results: The TCSPC approach showed decreases in mean time-of-flight and in time-integrated transmittance during the stimulation period. The temporal profile of time-resolved transmittance is influenced by the absorption and scattering 2coefficients as well as by their spatial distribution within the sampled volume. For a quantitative analysis of the observed changes, TCSPC experiments on suitable phantoms and numerical simulations are needed to deduce possible changes of the absorption and scattering properties during brain activation. The whole spectrum approach showed changes in light attenuation over the whole spectral range, correlated to the stimulation period. The time course of these changes differed over the spectral range. Discussion & Conclusions: In this study 0 ) we demonstrated that two new NIRS approaches improve the spectral and the temporal resolution of the method. Changes in scattering may contain additional information about physiological cortical ~ processes (Gratton 1995). The whole spectrum approach may allow for better detection of cytochrome-oxidase. Also~ following the theory of Matcher (Matcher et al. 1994), pathlength changes can be calculated on the basis of the water .___ absorption peak in the NIR region. -~ We conclude that both new NIRS techniques are o sensitive enough to detect haemodynamic changes in the cortex in response to functional cortical stimulation. Additional information is obtained, since scattering properties and the full NIR spectrum are monitored by the respective approach. Literature: run time of experiment / s 0,896