Three-dimensional Superposition of Diffuse Optical Tomography Results and Subjacent Anatomic Structures Christina Habermehl 1 , Christoph Schmitz 1,2 , Jan Mehnert 1,3 , Susanne Holtze 1,3 , Jens Steinbrink 1,4 1 Berlin NeuroImaging Center,Charité Dept. of Neurology, Charitépl. 1, 10117 Berlin, Germany, 2 NIRx Medizintechnik GmbH, Berlin, Baumbachstr. 17, 13189 Berlin, Germany, 3 Max Planck Institute for Human Cognitive and Brain Sciences, Stephanstr. 1a, 04103 Leipzig, Germany, 4 Center for Strokeresearch Berlin (CSB), Charité Universitätsmedizin, 10098 Berlin, Germany ABSTRACT Near infrared spectroscopy (NIRS) and diffuse optical tomography (DOT) reveal no information about the measurement’s underlying anatomical structures. An independent anatomical mapping of DOT results onto the subject’s brain or a generic brain model is desirable, especially when regions prone to large inter-subject variability are studied. We show two methods to match DOT data from high density fiber grids to anatomical structures. The forward model that is used to predict the light propagation is based on one generic anatomical MR scan. In both approaches we use this model MR-scan to translocate the position of the optical fiber grid from our experimental setup to the FEM model space. The fiducial mark approach uses the spatial normalization of the subject’s MR-scan (with marked corners of the fiber grid) and the model’s MR scan, leading to a translocation of the fiber pad position to the FEM-Model space. The anatomic landmark approach is used without individual MR scan. 19 reference points and the position of the fiber pad corners are determined using photogrammetry software. These coordinates are translocated to the FEM model space by solving the least square problem of the subject’s and the model’s reference points. We illustrate and compare both methods and show results from a vibrotactile stimulation experiment in humans. Keywords: near infrared spectroscopy, diffuse optical tomography, anatomical mapping, human brain function 1. INTRODUCTION Non-invasive NIRS is an established tool for revealing cortical activation through changes in the degree of hemoglobin oxygenation in the activated area. Near infrared light is given into the subject’s brain by optical fibers, which are placed on the head surface. Differences in the light attenuation are then used to calculate interiour absorption changes that are mainly caused by a cortical activation and a locally increased blood flow. The topographic approach in NIRS, uses data from next nearest neighbours with source-detector-distances of 2-4cm. This method is an established tool in phyiological and psychological research [1-3] and it is known to have a good temporal but a rather poor lateral resolution. The lateral resolution can be enhanced by a multi-distance approach [4, 5]. Arrays of closely spaced fibers allow the simultaneous measurement of light paths from different tissue depths. These overlapping photon paths and an image reonstruction procedure lead to a series of three-dimensional reconstructed images of changes of interiour optical properties. This optical tomography approach has a rough depth discrimination and facilitates the separation of signals from superficial and deeper layers. The image reconstrution procedure for 3D DOT contains of two major steps. First, the light propagation in the medium with asumed optical properties is modeled. This leads to a sensitivity matrix that contains information about the contribution of each voxel in the medium to the measured surface data. Secondly, the inverse problem of recovering interior optical properties from the surface data is solved, mainly by multiplying the inverted sensitivity matrix and the detected light intensity values from the surface. Image reconstruction is an ill-posed and under-determined problem and many groups work on one or both parts of it [5-11].