Matthias F. Valverde Salzmann

Alumni Department Physiology of Cognitive Processes
Alumni of the Department High-Field Magnetic Resonance

Main Focus



New collaborations arised from an exhibition of my old polarized light microscopy data (more details @ 2012 below). So we are currently processing some more post-mortem human tissue samples for further investigations.

Here are some more examples:

Figure "FiberWheel": Sections with myelinated parallel fibers arranged in radial and tangential orientations to illustrate color coding.

Figure "XTracts": Two sections with myelinated parallel fibers arranged one upon the other in different orientations. Vector averaging artifacts in the overlapping region.

Figure "Callithrix": Coronal cross section through the cortical hemisphere of a common marmoset.

Figure "MotorTracts": Motor cortex of a macaque monkey.

Figure "HumanTracts": Motor cortex.


Concurrent ultra-highfield fMRI @14T and optical imaging spectroscopy in rats and new world monkeys (Callithrix jacchus).

M.F. Valverde Salzmann, R. Pohmann


Early functional studies on the primary visual cortex of primates and cats have revealed a striking topological organization of functional domains for orientation selectivity, color processing and other visual features [1-5]. Anatomical investigations seemed to suggest that the functional columnization (subdivision into columns) of V1 and other cortical areas is primarily based on cytoarchitectonic principles [6]. Over the last years - especially due the increased interest in the origin of the BOLD signal- the notion that the angioarchitecture and specific mechanisms of local blood flow regulation might fundamentally contribute in generating the functional 'columnar' organization of the visual cortex, has gained importance. However, investigations on the functional role of the vascular networks, especially the microcirculatory networks within cortical columns are difficult since they require the detection of the course of thousands of thin (~5µm) capillary filaments along the three dimensions of a cortical column and its immediate vicinity or even over larger cortical regions (ideally > 2 mm3) [7-9]. Current functional imaging technologies like MRI, 2-Photon microscopy, optical imaging of intrinsic signals or voltage sensitive dyes, do either lack the special resolution that is necessary to resolve the fine structure of the angioarchitecture or have limited tissue penetration depth and thus restrict investigations to thin regions between the cortical surface and a few hundred microns below it. Thus anatomical techniques such as histological stains and high resolution Micro-CTs, in combination with functional imaging approaches have to be applied to a small brain region containing several cortical columns in order to gain an overview over all relevant components. Only by such multimodal imaging datasets virtual 3D reconstructions can be gained on which then modelling of the functional principles of cortical blood supply mechanisms can be done.

Figure: Angiography and functional columns. (A) micro-CT angiography of intracortical vessels. (B) animal model and areas of interest. (C) comparision of vascular structure graph and functional columns (V1, marmoset, intrinsic optical imaging).


The aim of our project is to gain deeper insights into the regulatory mechanisms of cortical blood flow, in particular regarding the supply of a single functional column. We think that understanding the functional column might be a cornerstone for a better understanding of the BOLD signal.


In order to achieve this goal, we designed a novel multimodal imaging system that enables us to perform concurrent ultrahighfield fMRI @ 14 Tesla and optical imaging spectroscopy in rodents and small New World monkeys. In addition we are running collaborations with the Institute of Inorganic Chemistry and the Department of Preclinical Imaging and Radiopharmacy at the University of Tübingen in order to develop multimodal contrast agents based on lanthanide-based nanocrystals that enable us to image the vascular microarchitecture in functional and anatomical MRI experiments as well as in post-mortem high resolution micro-CTs (see below for details). Histological processing of the tissue specimen will complement the methodological framework that will allow us to gain virtual reconstructions of the angio- as well as the cytoarchitecture.

Figure: Schematic representation of the rat fMRI/optical imaging setup components. A tandem lens system is used to project images from the cortex onto the camera sensor. The camera itself is located outside the gradient. A right angle prism bends the image path. Light is guided to the brain by optical fibers. The RF-coil is located under the lower prism face.

Initial results

  • In cooperation with a german manufacturer of high end scientific cameras we developed a camera systems that allows us to perform optical imaging spectroscopy of intrinsic signals at magnetic field strengths of 14 Tesla and thus to perform concurrent fMRI and optical imaging inside our small animal 14T MRI scanner. Our initial experiments (see below) give proof-of-concept for our new multimodal imaging system.

Figure: Comparison of intrinsic signals vs BOLD signals recorded concurrently in a rat motor cortex (S1) during an fMRI experiment @ 14 Tesla. Upper row: Intrinsic signal maps @ 630nm (Red), @ 540nm (Green) and the corresponding BOLD signal map. Lower row: Layer dependent BOLD signal distribution. Stimulus: forepaw region, 27 pulses were presented at variing frequencies. Right: Timecourses of the intrinsic signal and BOLD signal.

  • In cooperation with Prof. Dr. Hermann A. Mayer and Farhad Jafarli (Institute of Inorganic Chemistry) we successfully developed a methodological protocol to produce hybrid nanoparticles made up by BaYbF5 nanocrystals that are encapsulated within a silica shell. Testing of the novel contrast agents was performed with the help of Prof. Dr. Bernd J. Pichler and Julia G. Mannheim (Department of Preclinical Imaging and Radiopharmacy). Currently, further modifications for surface functionalization of the nanoparticles for MRI and histological detection are carried out.
  • We previously developed methodological protocols for 3D-reconstructions of histologically processed cortical tissue that facilitates the alignment of functional and anatomical datasets and increases accuracy of such alignments [9-10].
  • High resolution micro-CT tomographic data could be gained from several specimen (rats) using a barium sulfate suspension as contrast agent. High resolution 3D reconstructions of the angioarchitecture were performed. This data was then used as a backbone for the development of our vascular data analysis software tool: the Open Vascular Network Explorer (alpha version) that allows virtual exploration and blood flow simulations on vascular structure graphs.

Initial conclusion

A technical and methodological framework for high-end functional and anatomical investigations of the origin of the hemodynamic response has been set up, consequently enabling us over the course of the project to generate models of the fundamental principles of the regulation of cerebral blood flow within a single functional column.

Figure: Micro-CT - 3D reconstruction of intracortical vasculature within a cylindrical tissue sample (4mm radius).


[1] Grinvald, A., et al. (1986). "Functional architecture of cortex revealed by optical imaging of intrinsic signals." Nature 324(6095): 361-364.

[2] Bonhoeffer, T. and A. Grinvald (1991). "Iso-orientation domains in cat visual cortex are arranged in pinwheel-like patterns." Nature 353(6343): 429-431.

[3] Landisman, C. E. and D. Y. Ts'o (2002). "Color Processing in Macaque Striate Cortex: Relationships to Ocular Dominance, Cytochrome Oxidase, and Orientation." J Neurophysiol 87(6): 3126-3137.

[4] Valverde Salzmann, M. F., et al. (2012). "Color Blobs in Cortical Areas V1 and V2 of the New World Monkey Callithrix jacchus, Revealed by Non-Differential Optical Imaging." The Journal of Neuroscience 32(23): 7881-7894.

[5] Vanzetta, I. and A. Grinvald (2008). "Coupling between neuronal activity and microcirculation: implications for functional brain imaging." HFSP Journal 2(2): 79-98 [1] Blinder, P., et al. (2013). "The cortical angiome: an interconnected vascular network with noncolumnar patterns of blood flow." Nat Neurosci 16(7): 889-897.

[6] Lund, J. S., et al. (2003). "Anatomical Substrates for Functional Columns in Macaque Monkey Primary Visual Cortex." Cerebral Cortex 13(1): 15-24.

[7] Blinder, P., et al. (2013). "The cortical angiome: an interconnected vascular network with noncolumnar patterns of blood flow." Nat Neurosci 16(7): 889-897.

[8] Hirsch, S., et al. (2012). "Topology and hemodynamics of the cortical cerebrovascular system." J Cereb Blood Flow Metab 32(6): 952-967.

[9] Valverde Salzmann, M. F., et al. (2011). "Multimodal vessel mapping for precise large area alignment of functional optical imaging data to neuroanatomical preparations in marmosets." Journal of Neuroscience Methods 201(1): 159-172.

[10] Valverde Salzmann, M. F., et al. (2011). "High-resolution imaging of vessels in the isolated rat brain." ISMRM: Montreal, Quebec, Canada.


The  'Concurrent fMRI and Optical Imaging Spectroscopy' project continues...

Figure: Ex-vivo microCT of a rat head. Large blood vessels visualized by filling with radiocontrast agent.


From January until December 2013, I was involved in two projects:

1. In collaboration with Prof. Dr. Klaus Scheffler and Dr. Rolf Pohmann (department of high-field magnetic resonance) I designed an experimental setup for concurrent fMRI and optical imaging spectroscopy (OIS) in rats (at 14 Tesla). This project aims at gathering new insights into the hemodynamic mechanisms underlying the BOLD signal.

Figure: CAD sketch of a prototype fMRI/OIS setup design.

2. As part of my collaboration with Prof. Dr. Nikos Logothetis and Dr. Henry Evrard (department of physiology of cognitive processes), I introduced the software AMIRA (FEI) for automatic registration of functional and anatomical MRI datasets to a 3D brain atlas.

Figure: 3D macaque monkey brain atlas (software: Amira, VSG);


In 2012, I worked as a postdoc for Dr. Axel Thielscher on his TMS-research project. For more information follow this . Aim of this study was -and still is- to investigate the myeloarchitecture of the human motor cortex, in order to gain detailed insights into its anatomical fine structure which can be used for accurate modeling of the cortical tissue.

Beside histological staining approaches we use polarized light microscopy to trace the course of fibers in gray and white matter.

Since the beginning of 2013, these investigations are continued in the form of an ongoing collaboration.

Figure: Myeloarchitecture of a white matter/gray matter region in the visual cortex of a primate. The vector field map of fiber tract orientations shown here was calculated from a serie of microscope images (of a single unstained histological section) taken at different pol-filter azimuth direction angles. Left: Color coded fiber tract orientations. Right: A maximum intensity projection of the pol-image serie.


Project finished Dec. 2011.

Footage: Intrinsic optical imaging of orientation columns in primate V1 (common marmoset; single condition maps).


Functional-anatomical investigations on the visual system in New World Monkeys.

M.F. Valverde Salzmann, A. Schüz


Functional studies on the visual cortex of primates has revealed a striking topology of domains for orientation selectivity as well as domains for color processing [1,2,3]. Anatomical studies indicate that this functional segregation is fundamentally based on the cytoarchitecture of the visual system. Many studies investigated the relationship between functional domains and architectonical structures such as cytochrome oxidase (CO) blobs. However, diverging results led to an ongoing controversy about the spatial relations of functional columns and thalamic input zones especially with respect to their tripartite subdivision into konio, parvo, and magno streams, associated with the perception of color, form, and motion [4,5].


The aim of our study is to investigate the arrangement of orientation columns on functional maps of orientation selectivity, with respect to the location of functional color domains and thalamic input zones such as CO blobs in V1 and V2 of the primates visual cortex.


We use intrinsic optical imaging, laminar microelectrode arrays and a variety of visual stimuli as achromatic gratings and color flicker to map the organization of functional domains in V1 and V2 of the common marmoset. Marmosets like other New World Monkeys show a well investigated genetic dichotomy that affects the expression of retinal photoreceptors sensitive for middle to long (M,L) wavelengths as well as the cellular basis of the parvo stream. Thus most male and some female marmosets are colorblind. However, few females show full trichromatic color vision [6,7]. This makes the marmoset an important model for investigations on the functional segregation of the three visual pathways.

Initial results

We developed a histological protocol for making 3D-reconstructions of the histologically processed cortex that facilitates the alignment of functional and anatomical data (also in Magnetic Resonance Imaging studies [8]) and increases accuracy. Thus, dislocation errors in alignments are reduced by more than 50% in comparison to those reported in previous studies [9]. We further developed a stimulus protocol that allows us to map color domains by using optical imaging in trichromatic marmosets. Our results show that color domains are most effectively activated by using red-green (L-M) flicker stimuli. We also showed that color domains in marmosets are colocalized with CO blobs in V1 and thin stripes in V2 [10].

Initial conclusion

We conclude that the observed color domain activation is primarily triggered by a modulation of the parvo stream ((L-M)-cone axis), thus supporting the notion that CO blobs in New World as well as in Old World trichromatic primates are segregated domains of color processing.


[1] Livingstone, M., D. Hubel: Anatomy and physiology of a color system in the primate visual cortex. The Journal of Neuroscience 4, 309-356 (1984).

[2] Landisman, C.E., D.Y. Ts’O: Color Processing in Macaque Striate Cortex: Relationship to Ocular Dominance, Cytochrome Oxidase, and Orientation. Journal of Neurophysiology 87, 3126-3167 (2002).

[3] Lu, H.D., A.W. Roe: Functional Organization of Color Domains in V1 and V2 of Macaque Monkey Revealed by Optical Imaging. Cerebral Cortex 18, 516-533 (2008).

[4] Livingstone, M., D. Hubel: Segregation of form color, movement, and depth: anatomy, physiology, and perception. Science 240, 740-9 (1988).

[5] Sincich, L.C., J.C. Horton: The Circuitry of V1 and V2: Integration of Color, Form, and Motion. Annual Review of Neuroscience 28, 303-26 (2005).

[6] Blessing, E.M., S.G. Solomon, M. Hashemi-Nezhad, B.J. Morris, P.R. Martin: Chromatic and spatial properties of parvocellular cells in the lateral geniculate nucleus of the marmoset (Callithrix jacchus). The Journal of Physiology 557, 229-245 (2004).

[7] Conway, B.R.: Color Vision, Cones, and Color-Coding in the Cortex. The Neuroscientist 15, 274-290 (2009).

[8] Valverde Salzmann, M.F., N.K. Logothetis, R. Pohmann: High-resolution imaging of vessels in the isolated rat brain. ISMRM: Montreal, Quebec, Canada (2011).

[9] Valverde Salzmann, M.F., D.J. Wallace, N.K. Logothetis, A. Schüz: Multimodal vessel mapping for precise large area alignment of functional optical imaging data to neuroanatomical preparations in marmosets. Journal of Neuroscience Methods 201, 159-172 (2011).

[10] Valverde Salzmann, M.F., A. Bartels, N.K. Logothetis, A. Schüz: Color blobs in cortical areas V1 and V2 of the new world monkey Callithrix jacchus, revealed by non-differential optical imaging. Journal of Neuroscience 32(23), 7881-7894 (2012).


Figure: Functional-anatomical investigations on the primary visual cortex in marmosets. (A) Intrinsic optical imaging (IOI) orientation map superimposed onto the 3D reconstruction of histological sections stained for cytochrome oxidase. (B) The pattern of the superficial cortical vasculature (from the same cortical region) labeled with FITC. Scale bar = 500 µm.

For more information see my journal article


Figure: Color domains in the visual cortex (left hemisphere) of 3 trichromatic marmosets (M1-M3). The figure show single condition maps of the intrinsic response to color varying (red/green) 1.5 Hz flicker stimuli and luminance varying (black/white) 1.5 Hz flicker. The chromatic condition maps reveal the typical regular patch pattern of functional color domains in V1 and V2 of primates.

For more information see my journal article


Imaging data resulting from a collaboration with :

Ultra-high resolution imaging of the cerebral vasculature in rats using a 16.4 Tesla MRI scanner.

Figure: Alignment of functional signals, anatomical MRI data and histological preparations by using the multimodal vessel mapping approach (IOI = intrinsic optical imaging, WM= white matter, CytOx = Cytochrome Oxidase).


Figure: Angioarchitecture of a rat brain. 3D reconstruction of the cerebral vasculature, visualized by using gadolinium based contrast agents and a 16.4 Tesla MRI scanner. a) 2 anatomical cross sections (horizontal [olfactory bulb on the left] and coronal [right]) and a volumetric surface plot of the vascular data. b+c) Coronal view of the vasculature. The intracortical vessels are oriented radially in the brain, resulting in this hedgehog-like appearance.

For more information see my poster


Imaging data recorded in collaboration with and Julia Mannheim at the Department of Preclinical Imaging and Radiopharmacy at the Faculty of Medicine at the University Hospital of the Eberhard Karls University of Tübingen.

Figure: Ex-vivo X-ray Computed Tomography of a common marmoset and a rat skull using a small animal Micro-CT scanner (Siemens).

Resolution: ~100µm isotrop; Scale bar ~ 10mm. 3D Reconstruction: Amira (FEI), Meshlab.

Curriculum Vitae

Please contact me to recieve an updated CV.

Go to Editor View