Virtual Histology with MRI - Brain Tissue Characterization 

Our brain is a complex system that includes many types of tissues, component and structures. Since the seminal works of Ramon y Cajal (1), Brodmann (2) and others (3-6) at the beginning of the last century, the brain's macro and micro structures were studied by dozens of histological staining procedures (1,2,4) (cyto-architecture and myelo-architecture). For example, Brodmann has used cyto-architecture mapping of the human cerebral cortex to differentiate between as many as 47 cortical areas (2) that later turned out to be also distinct in functional output. Brodmann's work as well as the histological studies of many others (7-9) are the basis of human and rodent brain atlases that are often used today in text-books and in neuroscience research (including neuroimaging and fMRI).

Magnetic resonance imaging (MRI) has the potential to provide structural information on neuronal systems due to its high sensitivity and resolution capabilities. This is the basis of neuro-radiological diagnosis of neuronal tissue by neuro-radiologist who define abnormal brain structures by qualitative, visually guided analysis. With the development of new image processing algorithms, the field of tissue characterization by neuroimaging in general and MRI in particular has evolved tremendously. Basic tissue segmentation and volumetric analysis are being used more frequently in neuroimaging research as well as in the clinics. For example, Figure 1 shows an axial slice of a T1-weighted MRI scan of a healthy subject along with its segmentation into gray matter, white matter and CSF components. Such gross tissue segmentation is can be achieved fairly easily by a handful of freeware software (partial list given in (10-14)). The vast majority of these algorithms use a single contrast MRI data-set.

One of MRI's greatest advantages is being a multi-modality technique, i.e. it is able to produce a multitude of image contrasts (15). MRI's relaxation times, T1 and T2, are the most basic contrast mechanisms in MRI. In addition, physical and biochemical properties of the tissue such as diffusion, macromolecular concentration, magnetic susceptibility and density can contribute to image contrast. We recently developed a method named imaging in which tissue segmentation can done automatically, region wise, based on multi-dimensional MRI data (16). This methodology combines basic image and signal processing steps including contrast enhancement, principle component analysis and clustering.

Using 10 different MRI contrast mechanisms were able to segment the thalamus into at least 7 sub-nuclei masses (Figure 2). Each color in the thalamus region represents a significantly different contrast cluster. Comparison with atlases provided assignment of these clusters to known thalamus sub-nuclei: pulvinar, medio-dorsal, ventral anterior, ventral-posterior lateral/medial, lateral-posterior/dorsal, anterior and ventral lateral. With different combinations of MRI contrasts utilizing the inversion recovery pulse sequence, we were able to segment the cortex into laminar sub components (Figure 3). Inversion recovery (IR) is one of the basic magnetic resonance pulse sequences traditionally used to measure the longitudinal relaxation time, T1. One variable of the IR method is the inversion time which, if tuned properly, can achieve unique image contrast as it can zero the signal of specific tissues. Figure 3 shows IR-MR images taken at different TIs. As TI increases, the contrast between different parts within the cortex becomes visible. The different IR images, reveal a laminar pattern along the cortex as indicated by the multi-spectral cluster analysis (Figure 3).


1.Ramón y Cajal S. Elementos de Histología Normal y de Técnica Micrográfica. Madrid: Editorial Pascual Aguilar; 1897
2. Brodmann K, Garey L. Brodmann's Localisation in the cerebral cortex. London, River Edge, NJ: Imperial College Press; 1999
3. von Economo CB, Koskinas GN. The Cytoarchitectonics Of The Adult Human Cortex. Vienna and Berlin: Julius Springer Verlag; 1925
4.Paxinos G, Mai JrK. The human nervous system. San Diego , Calif. ; London: Elsevier Academic Press; 2004. xvii
5. Kandel ER, Schwartz JH, Jessell TM. Principles of neural science. New York: McGraw-Hill, Health Professions Division; 2000
6. Finger S. Origins of neuroscience : a history of explorations into brain function. New York ; Oxford: Oxford University Press; 1994
7. Martin JH. Neuroanatomy : text and atlas. New York ; London: McGraw-Hill; 2003.
8. Paxinos G, Watson C. The rat brain in stereotaxic coordinates. Amsterdam ; Boston: Elsevier Academic Press; 2005.
9. Amunts K, Zilles K. Advances in cytoarchitectonic mapping of the human cerebral cortex. Neuroimaging Clin N Am 2001;11(2):151-169
10. Ashburner J, Friston K. Multimodal image coregistration and partitioning--a unified framework. Neuroimage 1997;6(3):209-217.
11.Wu DH, Chen AD, Johnson CS. An Improved Diffusion-Ordered Spectroscopy Experiment Incorporating Bipolar-Gradient Pulses. Journal of Magnetic Resonance, Series A 1995;115(2):260-264
12. Fischl B, Salat DH, Busa E, Albert M, Dieterich M, Haselgrove C, van der Kouwe A, Killiany R, Kennedy D, Klaveness S, Montillo A, Makris N, Rosen B, Dale AM. Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron 2002;33(3):341-355
13. Atkins MS, Mackiewich BT. Fully automatic segmentation of the brain in MRI. IEEE Trans Med Imaging 1998;17(1):98-107
14.Stokking R, Vincken KL, Viergever MA. Automatic morphology-based brain segmentation (MBRASE) from MRI-T1 data. Neuroimage 2000;12(6):726-738.
15. Stark DD, Bradley WG. Magnetic resonance imaging. St. Louis ; London: Mosby; 1999.
16. Yovel Y, Assaf Y. Virtual definition of neuronal tissue by cluster analysis of multi-parametric imaging (virtual-dot-com imaging). Neuroimage 2007;35(1):58-69

Image Gallery

Contact Us

Our lab is located at Sherman building, room #421. This building is part of the George S. Wise faculty of life-sciences, Tel Aviv University

Address: Sherman Bldg, room #421

Tel. (Yaniv):+972-3-6407661

Tel. (Lab):+972-3-6406980


E-mail: assafyan -at-