**4. Three-dimensional electronic atlases based on histological data**

The digitization of previous published atlas led to problems in accuracy due to errors inherent in the technique used to construct them. In order to achieve better precision and accuracy in the 3D atlases, three groups have proposed to generate their electronic threedimensional reconstructions based on own histological sections instead of aligning and trying to correct previous published templates.

#### **4.1 The creation of a brain atlas for image guided neurosurgery using serial histological data (Chakravarty et al, 2006)**

Until 2006, the group from MNI (Montreal Neurological Institute, Canada) used the atlas developed by St-Jean et al., 1998 to program neurosurgical interventions. However, some shortcomings were recognized, including limited inherent resolution in the slice direction, limited number of structures, and some small mis-registrations between the digital atlas and the Colin27 MRI average that are propagated to patient MRI data during the atlas customization procedure. In this manuscript, the authors addressed these limitations and presented a technique for the creation of a brain atlas of the basal ganglia and thalamus derived from serial histological data. The technique used was identical to the one used in St-Jean's (1998) atlas. However, in the latter instance own histological preparations were available instead of scanned figures from the Schaltenbrand & Wahren atlas. The authors digitized coronal histological sections and delineated 105 anatomical structures in them. A slice-to-slice nonlinear registration technique to correct for spatial distortions was

Review of Printed and Electronic Stereotactic Atlases of the Human Brain 161

Orientation and interval of sections

Coronal sections (0.70 mm interval)

coronal sections

(0.9 or 1.0 mm) intervals Range of Slices

Thalamus, hypothala mus, basal ganglia amygdala and hippocamp us

> Basal ganglia

Thalamus, basal ganglia, subthalamic fiber tracts

Coordinate System

> ICP MSP VAC

ICP MSP

ICP MSP VPC Correction of tissue deformation

The ANIMAL slice-toslice registration procedure is applied

Automated Processing for data coregistration and semiautomatic processing

Calculated distortion factors from ICL mesurement

Registratio n and atlas-topatient normalization

2 Steps ANIMAL application

BALADIN algorithm

Not described

Methods Staining and

other methods

Luxol Blue (myelin)

Nissl(cell Bodies)

80 sections Nissl-stained ; 80 sections immunostaini ng for calbindin

Nissl ; Myelin; Calciumbinding proteins(nonphosphorylated neurofilament protein and Acetylcholine sterase

Table 3. Three-dimensional electronic atlases based on histological data

**4.4 The São Paulo-Würzburg electronic atlas of the human brain initiative** 

Since 2005, the Brain Bank of the Brazilian Aging Brain Research Group (BBBABSG) of the University of São Paulo Medical School (USPMS) collaborates with researchers interested in neuroimaging-neuropathological correlation studies including dementias, white matter hyperintensities, and epilepsy. As a particularity, the BBBABSG is linked to the MRI section of the USPMS, so the brains can be scanned postmortem in-situ within a short postmortem

At the Julius-Maximilian University of Würzburg (Germany), a fast, reliable, and easy to use celloidin method for serial sections of the human brain has been developed (Heinsen et al.,

Reference Nr of

Chakravarty et al., 2006

Yelnik et al, 2007

Krauth et al.,2010

interval.

2000).

brains and Sections

1 block from the left hemisphe re (86 pairs of slices)

1 Brain

800 sections

3 Brains (6 stacks), frozen, cut at 40- 50 µm

Mounted on paraffin; 105 anatomical structures manually delineated

Frozen sections, 70µm thick; Post mortem MRI

Postmorte m MRI (3 stacks);algo rithm "bootstrap approach" was used to construct an 3D average

introduced into the histological data set at the time of acquisition. Since the histological data were acquired without any anatomical reference, this registration technique was optimized to use an error metric which calculates a nonlinear transformation minimizing the mean distance between the segmented contours between adjacent pairs of slices in the data set. To register the atlas to Colin27, a pseudo-MRI was created by setting the intensity of each anatomical region defined in the geometric atlas to match the intensity of the corresponding region of the reference MRI volume. This allowed the estimation of a 3D nonlinear transformation using a correlation-based registration scheme to fit the atlas to the reference MRI. The result of this procedure was a contiguous 3D histological volume, a set of 3D objects defining the basal ganglia and thalamus, both of which are registered to a standard MRI data set, to be used for neurosurgical planning.

#### **4.2 A three-dimensional, histological and deformable atlas of the human basal ganglia. I. Atlas construction based on immunohistochemical and MRI data (Yelnik et al., 2007)**

The French group describes in detail the construction of an atlas of the human basal ganglia. One brain was selected for reconstruction, and prior to the removal from the skull it was subject to MRI acquisition and then cryosectioned. The MRI was used for the coregistration of the atlas and permitted the production of more consistent 3D surfaces. Three different software tools were used in this study. A Threedimensional Tracing Software application (TTS) was developed for the purpose of digitization and processing of serial cerebral contours. The second tool is called Yav++ and its principal features include comparison and fusion of 3D images in multiplanar viewers as well as a 3D camera allowing visualization of serial contours, 3D surfaces, and 3D images in the same image. The last software tool is called BALADIN software, and it is an automatic image registration algorithm that allows registration of 2D or 3D images through an intensity-based block-matching approach. The novelty of this atlas is the MRI acquisition, which represents the core data element of the study.

#### **4.3 A mean three-dimensional atlas of the human thalamus: Generation from multiple histological data (Krauth et al., 2010)**

The stereotactic anatomical atlas under consideration consists of a three-dimensional thalamic model derived of a mean from six series of histologically processed brain sections. Postmortem MRIs are available for three of the stacks. The authors recommend that atlases should be based on a population instead of individuals. Therefore, previously studied stacks (Morel, 2007) were reconstructed based on multiarchitectonic criteria and integrated by an iterative algorithm -the so called bootstrap approach- to result in a three-dimensional average from three brains. The authors contend that their atlas improves the previous work on thalamic model reconstruction in several aspects. Firstly, while those models are based on the geometry seen in a single stack, their model incorporates topological and geometric details from different stacks and different stereotactic directions. Secondly, it would represent the average anatomy of several specimens instead of a single one, which would remove the bias towards a specific individual. This paper describes mainly how previous stacks can be used to build an average model of the human thalamus and the advantages of this kind of approach.

introduced into the histological data set at the time of acquisition. Since the histological data were acquired without any anatomical reference, this registration technique was optimized to use an error metric which calculates a nonlinear transformation minimizing the mean distance between the segmented contours between adjacent pairs of slices in the data set. To register the atlas to Colin27, a pseudo-MRI was created by setting the intensity of each anatomical region defined in the geometric atlas to match the intensity of the corresponding region of the reference MRI volume. This allowed the estimation of a 3D nonlinear transformation using a correlation-based registration scheme to fit the atlas to the reference MRI. The result of this procedure was a contiguous 3D histological volume, a set of 3D objects defining the basal ganglia and thalamus, both of which are registered to a standard

**4.2 A three-dimensional, histological and deformable atlas of the human basal ganglia. I. Atlas construction based on immunohistochemical and MRI data (Yelnik et** 

The French group describes in detail the construction of an atlas of the human basal ganglia. One brain was selected for reconstruction, and prior to the removal from the skull it was subject to MRI acquisition and then cryosectioned. The MRI was used for the coregistration of the atlas and permitted the production of more consistent 3D surfaces. Three different software tools were used in this study. A Threedimensional Tracing Software application (TTS) was developed for the purpose of digitization and processing of serial cerebral contours. The second tool is called Yav++ and its principal features include comparison and fusion of 3D images in multiplanar viewers as well as a 3D camera allowing visualization of serial contours, 3D surfaces, and 3D images in the same image. The last software tool is called BALADIN software, and it is an automatic image registration algorithm that allows registration of 2D or 3D images through an intensity-based block-matching approach. The novelty of this atlas is the MRI acquisition, which represents the core data element of the

**4.3 A mean three-dimensional atlas of the human thalamus: Generation from multiple** 

The stereotactic anatomical atlas under consideration consists of a three-dimensional thalamic model derived of a mean from six series of histologically processed brain sections. Postmortem MRIs are available for three of the stacks. The authors recommend that atlases should be based on a population instead of individuals. Therefore, previously studied stacks (Morel, 2007) were reconstructed based on multiarchitectonic criteria and integrated by an iterative algorithm -the so called bootstrap approach- to result in a three-dimensional average from three brains. The authors contend that their atlas improves the previous work on thalamic model reconstruction in several aspects. Firstly, while those models are based on the geometry seen in a single stack, their model incorporates topological and geometric details from different stacks and different stereotactic directions. Secondly, it would represent the average anatomy of several specimens instead of a single one, which would remove the bias towards a specific individual. This paper describes mainly how previous stacks can be used to build an average model of the human thalamus and the advantages of

MRI data set, to be used for neurosurgical planning.

**histological data (Krauth et al., 2010)** 

this kind of approach.

**al., 2007)** 

study.


Table 3. Three-dimensional electronic atlases based on histological data

#### **4.4 The São Paulo-Würzburg electronic atlas of the human brain initiative**

Since 2005, the Brain Bank of the Brazilian Aging Brain Research Group (BBBABSG) of the University of São Paulo Medical School (USPMS) collaborates with researchers interested in neuroimaging-neuropathological correlation studies including dementias, white matter hyperintensities, and epilepsy. As a particularity, the BBBABSG is linked to the MRI section of the USPMS, so the brains can be scanned postmortem in-situ within a short postmortem interval.

At the Julius-Maximilian University of Würzburg (Germany), a fast, reliable, and easy to use celloidin method for serial sections of the human brain has been developed (Heinsen et al., 2000).

Review of Printed and Electronic Stereotactic Atlases of the Human Brain 163

anatomic variability. Consequently, errors in diagnosis and neurosurgical interventions impend on generalizing interrelationships of a single brain. Probabilistic maps can include features such as cytoarchitecture, chemoarchitecture, blood flow distributions, metabolic rates, behavioral and pathologic correlates, electrophysiologic tissue characteristics and others (Mazziota, 1995). Some previously described atlases list probabilistic features (Andrew et al., 1969; Afshar et al., 1978; Krauth et al*.,* 2010). In this paragraph, special focus

**5.1 A probabilistic atlas and reference system for the human brain: International** 

Through an International Consortium for Brain Mapping (ICBM) a data set that includes 7000 subjects between the ages of eighteen and ninety years and including 342 mono- and

Data on each subject include detailed demographic, clinical, behavioral, and imaging information. DNA has been collected for genotyping from 5800 subjects. A component of the program uses post-mortem tissue to determine the probabilistic distribution of microscopic cyto- and chemoarchitectural regions in the human brain. This can be combined with macroscopic information about structure and function derived from subjects *in vivo*, providing the an opportunity to gain meaningful insights into the concordance or discordance in micro- and macroscopic structure and function (Mazziota et al., 2001).

**5.2 A probabilistic functional atlas of the human subthalamic nucleus (Nowinski et al.,** 

The concept of probabilistic functional atlas (PFA) was introduced by Nowinski. His idea is to overcome limitations of the current electronic stereotactic brain atlases, such as anatomical nature, spatial sparseness, inconsistency, and lack of population information. The PFA is an algorithm that converts the coordinates of the neurologically most effective contacts into probabilistic functional maps, taking into account the geometry of a stimulating electrode and the patient's anatomy. Nowinski published the use of this algorithm to build an atlas of the subthalamic nucleus and of the ventrointermediate

This paper introduces a method for generation and validation of a probabilistic functional brain atlas of subcortical structures from electrophysiological and neuroimaging data. The method contains three major steps: (1) acquisition of pre-, intra-, and postoperative multimodal data; (2) selection of an accurate data set for atlas generation; and (3) generation of the atlas from the selected data set. The method is here applied to construct the probabilistic functional atlas of the human subthalamic nucleus from data collected during surgical treatment of 184 patients with Parkinson's disease. It is based on preoperative X-ray ventriculography imaging, intraoperative electrophysiological measurements and X-ray imaging, and postoperative neurological assessment. This method can be used to build

**5.3 A probabilistic functional atlas of the VIM nucleus constructed from pre-, intraand postoperative electrophysiological and neuroimaging data acquired during the** 

This work addresses construction of the PFA for the ventrointermediate nucleus (PFA-VIM). The PFA-VIM is constructed from pre-, intra- and postoperative electrophysiological and

**surgical treatment of Parkinson's disease patients (Nowinski et al, 2006)** 

will be on probabilistic atlases not classified in previous sections.

**consortium for Brain Mapping (ICBM)(Mazziotta et al., 2001)** 

thalamic nucleus (VIM) (Nowinski et al., 2004, 2006).

PFAs from other regions, as the next work from 2006 did.

dizygotic twins has been collected.

**2004)** 

*Cytoarchitectonic 3D contours of red nucleus (NR or RN, in red), subthalamic nucleus (STh in gray), and substantia nigra (in white) merged to the MRI from the same brain. PC in this case is the abbreviation for pedunculi cerebri. The question concerning the hypointense region in MRI antero-lateral to red nucleus is presently being studied in detail, applying this method to additional cases* 

#### Fig. 5. Cytoarchitecture x MRI

By combining both technologies, fundamental questions on post-mortem delay, appropriate fixation and neuroimagaing/neuropathological correlations could be addressed (Grinberg et al., 2008, 2009; Teipel et al. 2008).

The salient feature of this methodological approach to an atlas on the human basal ganglia is the post-mortem in situ MRI-scanning of the brain and the histological processing of the brain to generate serial 400 µm thick Nissl-stained sections. This protocol greatly facilitates cytoarchitectonic delineation of cortical and subcortical grey matter, compensation for shrinkage, and deformation and co-registration of high-resolution Nissl-stained sections with MRI scans.

Compelling results of match (red nucleus) and mismatch (subthalamic nucleus) of Nisslstained sections with the MRI-boundaries are depicted in Fig. 5.

The images are imported to software tools and warped to the original MRI scans, following nonlinear and linear correction protocols developed by the team and yet to be published. Three-dimensional cytoarchitectonic contours can then be compared to the original scans.All the illustrations presented in this chapter were made using our own material and techniques.
