Content uploaded by Armando Alaminos Bouza
Author content
All content in this area was uploaded by Armando Alaminos Bouza on May 05, 2020
Content may be subject to copyright.
Proceeding s of the Xlt h Meeting of the World Society for Stereota ctic and
Functi onal Neu rosurgery, Ixtapa, Mexico. Octo ber 11-15,1 993.
Stereota ct F un d Ncuro surg 1994 ;63:35-37
© 1994 S. Karger AG, Basel
1011-6 125/9 4/063 4-0035
S 8.00/0
C o m p ut e r W o r k s t a t io n - A s s ist e d S t e r e o t a x is
P.K. Pillay, P. Phavichtr
Department of Neurosurgery, Singapore General Hospital. Republic of Singapore
Com puter workstation-assisted stereotaxis was carried out on 107 patients over a period
of 1 year. Th e system used included a D EC 5000 workstation, optical disc drive, magnetic tape
drive. Truvel scanner, etheme t connections to MR I, and an X terminal in a dedicated stereo
tactic operatin g the ate r. The software was Stereoplan versions 1 and 2 (RS A. Boston) and uti
lized a Cosman-Roberts-Wells stereotactic frame. 54 ste reotactic craniotom ies for lesion
resection. 21 stereotactic biopsies, IS ste reotactic hemato ma aspirations and 14 functional
procedures were carried out.
The workstation was particularly useful for procedures involving volume stereotaxis.
Auto matic fiducial recognition, three-dim ensional (3-D ) volume rendering and multiple tra
jectory planning was rapid and accurate on the worksta tion. Stereoplan has the ability to aid
the surgeon in performing volumetric image-guided resections. This modality was particularly
useful in lesions with anatomically indistinct margins such as gliomas and in radiation neurosis.
We found the workstation to be less useful in point stereotaxis. The exception was in func
tional stereotaxis, where an anatomical 3-D model could be created of the thalamus, caudate
nucleus and cingulate gyrus.
The average time required to generate 3-D images, entry and target point coordinates
was 20.2 min (mean) on the workstation and about 7.3 min (mean) with the Radionics SCSI
laptop. The differences between target was 1 mm or less using both systems as a counterch eck.
A Ste r e o t a c tic S u rgic a l P la nn in g Sy s t e m f o r the IB M 3 86 / 48 6 P C Fa m ily
A. Alaminos, I. Ortega, H. Molina, P. Valladares
Cen tro Internacional de Rcstaturación Ncurológica. Ciudad de La Habana, Cuba
In 1990. after having done preclinical studies, we started the successful clinical applica
tion of the Neurorestorative Stereotactic Planning System (NSPS. Centro Internacional de
Resta turación Neurológica) in biopsies, cyst evacuations, fetal tissue tra nsplantation, selec
tive thalamotomies, de ep brain activity semimicrorecording, brachytherapy treatments with
l;2Ir intracavitary irradiation with :P and stereotactic radiosurgery.
The NSPS handles images from X ray. DSA, and M RI, but most of the work has been
done with CT images. Views of th e trajectories inside the brain can be displayed in axial, sag
ittal, coronal and arbitrary planes (e.g., probe ’s eye view). Reconstruction thickness can be
Downloaded by:
Stanford Univ. Med. Center
171.65.65.16 - 5/5/2020 5:39:22 PM
selected from one pixel to any desired value, in coronal and sagittal views; besides this, a te ch
nique was developed to emphasize the ventricular system in such a way that a ‘synthetic digi
tal' ventriculography can be displayed without contrast. Stereotactic coordinates and trajecto
ries remain valid indepen dent of the zoom state.
A subset of the Schaltenbrand and Wahren and Talairach atlases can be overlayed on the
slices after appropriate scaling.
Two versions of the programs are available, for Leksell and Riechert-Mundin ger stereo
tactic systems. No expensive or special hardware is needed; the minimum configuration can be
80386/80486 IBM PC compatible. 4 MB of RAM, 40 MB hard disk and 1 MB VG A video
controller.
3-D im e n sio n al Funct ional M ap p in g o f S e ns o rim o t o r C o rt e x by A c t iva te d
P ET S c annin g : C or r e l a t io n w i th M R a n d In tra o pe rati v e C o r t ica l S t im u la tio n
G.R. Cosgrove, B. Buchbinder, J. Jiang, W. Butler
Massachusetts General Hospital, Boston, Mass., USA
Localization of the central sulcus and the primary sensorimotor cortex is an important
consid eratio n for neurosurgical procedures in this region. We have developed a system for
integrated, 3-dimensional (3D) functional and anatomical mappin g of the cerebral cortex and
have explored the clinical utility and accuracy of this technique in 3 patients with structural
lesions (2 gliomas, I AVM) involving the primary sensorimo tor cortex. Cortical activation by
repetitive fist clenching was observed using a dynamic C |50 , inhalation PET rCB F technique.
Structural high-resolution 3D MR and gadolinium-enhanced 2D or 3D MR A were perform ed at
1.5 T. The PET, M R and MR A data sets were registered using a surface mating algorithm. Inte
grated 3D rendering of the cortical surface topography, cortical veins, primary lesion and site of
functional activation was performed. Intraoperative primary senso rim otormappingwas carried
out by direct cortical stimulation. In each case, the fist clenching motor task activated a localized
region in the contralateral pre- and postcentral gyri centered approxim ately on the central sul
cus. These localizations were concord ant with previous reportsof activation PET using the same
task in normal subjects. Localization of the primary sensorimotor cortex for the hand by activa
tion PE T was verified intraoperatively by direct cortical stimulation. 3D MR surface rendering
of cerebral topography and vascular anatom y was also examined at surgery and found to be in
excellent agreement with th e observed cortical anatomy. Integrated functional and anatomical
imaging of the cerebral cortex using activation P ET and 3D structural MR can accurately local
ize primary sensorimotor cortex and facilitate surgical procedures in this area.
Abstracts 36
Downloaded by:
Stanford Univ. Med. Center
171.65.65.16 - 5/5/2020 5:39:22 PM
A m yg da l o h ip p o c a m p a l A t r o ph y on M R I D o es N o t Pr e d i c t D ist r i b u t i o n of
Ict al O n s e t A re a in the Te m p o r a l Lo b e
T. Hoshidci, S. Uematsu, R.P. Lesser
Johns Hopkins Medical Institution, Baltimore. Md.. USA
We evaluated whether amygdalohippocampal complex atrophy seen on MRI can predict
the presence of ictal onset area (IOA) within the first 5 cm of anterior temporal lobe. Left to
right volume ratio of the amygdalohippocampal complex was used as the in dicator of atrophy
(a ratio of less than 0.9 indicates atro phy on the left side). The subdural grid covered the base
and lateral aspect of the left temporal lobe and a large area over the frontal and parietal lobes.
Twenty patients undergoing language-dominant left temporal lobectomy were studied.
MRI scan of the patients showed no appa ren t gross abnormality. In 14 of 20 patients, the
left amygdalohippocampal complex was atrophic; in th e remaining 6 patients there was no
apparent atrophy. In 8 of the 14 patients with atroph y (57%), IO A was localized in the an te
rior temporal lobe; in the remaining 6, IO A was extended to the posterior tem poral lobe and/
or th e extratemporal lobe. In only 1 of the 6 patients without atrophy (17%) was the IOA
localized in the anterior temporal lobe. In the remaining 5 patients, IOA was extended to the
posterior temporal lobe and/or extratemporal lobe (p = 0.119). These findings indicate that
amygdalohippocampal atrophy alone is not an indicator of the extent and distribution of IOA
in the temporal lobe and vicinity.
A Ne ur a l N e tw or k A n al y s is o f C S F Pr e s s u re Pa t t e r n s in N o r m a l P r e ssu r e
H y d ro c e p h a l u s
P. Mazzone", R. PisaniL. Forliinab, P. Arena1’, F. Nobilic
'Departm en t of Neurosurgery, E.O. Ospedali Galliera,
bCatania University - Electronic Dep artmen t and
cNeurophysiopathology, University o f Genova, Italy
28 patients underwent ventriculoperito neal shunt placement for the treatment of the so-
called normal pressure hydrocephalus (NP H). One of the most relevant indices for the diagnosis
and tr eatment of NPH was the CSF pressure patterns. The histograms of continuous pressure
monitoring were registered by a CMZ 801SICP Pressure Meter processor. A multilayer neural
network (Perceptrone) was used to study the pressure patterns by a back propagation method to
make an alternative classification of patients, based on CSF pressure histograms only.
The data founded by neural network examination indicate that (1) a difference exists
between classification made by the expert and the network, (2) it is possible th at these differ
ences depend on a small series of the 'patterns' or on a finding inside the pressure pattern s, and
(3) the classification of the neural network appears relevant to recognize the 'respo nd er'
patients to shunting procedures.
Abstracts 37
Downloaded by:
Stanford Univ. Med. Center
171.65.65.16 - 5/5/2020 5:39:22 PM
