156

D. Balériaux1)

1) Department of Neuroradiology, Hôpital Erasme, Brussels, Belgium.

Neuroradiology: past, present and future

One of the earliest radiographs ever made wasof the human skull, and by 1896 radiology hadproved itself in localizing bullets in the skull andcervical spine1,2. Plain skull radiographs wereof immense value in detecting trauma and thepresence of foreign bodies, but imaging of thebrain itself presented a seemingly insuperableproblem: the problem of visualizing the radio-lucent soft tissue of the brain within the virtuallyradiopaque box of the skull. In spite of the rapidadvances in X-ray imaging technology, it was tobe nearly 70 years before a satisfactory solutionwas found, allowing pathological changes to bevisualized in the tissue of the brain itself.

Indirect signsThe earliest methods used for the detection ofbrain tumours depended on indirect signs. Forexample, as the pineal gland is normallycalcified, it shows up on the X-ray image,marking the midline of the brain. A shift in itsposition indicates the presence of a space-occupying lesion. Infrequently, the tumour itselfmight also be calcified, allowing it to be observeddirectly (Fig. 1).

Other indirect signs were used for detectingintracranial hypertension. These includedthinning of the sella and, in children, digitalimpressions on the inside of the skull andopening of the cranial sutures, due to expansionof the brain.

Early detection of hypertension is important,as it allows appropriate measures to be taken toavoid neurological deterioration in children,while relief of hypertension in adults cansometimes restore demented patients to normalwithin a few days.

Indirect signs continued to play an importantrole in cranial diagnostics until well into the1960’s.

Contrast media Direct techniques for visualizing elements of thebrain, such as the ventricles and the blood

vessels, as well as the spinal cord, depended onthe use of contrast media.

As described by Dr O. Wayne Houser in hisPresident’s Address at the 1994 RSNA3, theidentification of suitable contrast media for usein neuroradiology was, to a certain extent, therecognition of the value of chance findings.

The first technique to appear was the use ofair for visualizing the outline of the brain andthe ventricular system (Fig. 2a), closely followedby direct injection of contrast media into thecerebral ventricles (Fig. 2b).

The former technique, first presented byDandy in 19184, remained in general use untilthe mid 1970’s. It depended on the fact that airwas much less absorbent to X-rays than theliquor normally present within the ventricles. It did not really show the brain itself, but only spaces within and around the brain.Nevertheless, compression or displacement ofthe ventricles was a useful indirect sign of thepresence of a tumour.

Although air was also used for delineation of the spinal cord (Fig. 4), it was quicklysuperseded by the use of positive contrastmedia, such as iodized oil or iophendylate(Pantopaque). This was the result of anaccidental discovery by Sicard and Forestier,who observed the flow of oil into the thecal sacfollowing an injection into the lumbarmusculature5.

Another important development was theintroduction of cerebral angiography, firstapplied in vivo by Egas Moniz, who injectedsodium iodide directly into the carotid artery6.Subsequently he was to use thorium dioxide7,which gave excellent short-term results, butproved to have long-term toxicity. In order tocapture the rapid flow of contrast mediumthrough the cerebral arteries, Moniz mountedhis films on a turntable, operating on theprinciple of a revolver. Later, multiple ex-posures were obtained using the ElemaSchönander AOT film changer.

157

There are two principal applications forcerebral angiography. The first is the detectionof lesions or abnormalities in the arteriesthemselves, such as occlusions, aneurysms orarteriovenous malformations (AVM’s). Thishas always been the most important application,but the advent of new imaging modalities hasresulted in a significant reduction in the totalnumber of cerebral angiographies performed.

The second application is the detection oftumours by observing displacement of thevessels or neovascularization. Neuroradiologistsrapidly learned to recognize the characteristicpatterns of the arteries and veins, and to detectany abnormalities which could indicate thepresence of a lesion. This is gradually becominga lost art.

Tomography and subtractionThe use of air and positive contrast mediashowed structures that were previously invisible,but they did not always show them very clearly.The faint images of the structures underexamination were often obscured by theoverlying bones of the skull and spinal column.

Two ingenious solutions to this problem weresuggested by B.G. Ziedses des Plantes in hisdoctoral thesis Planigraphy and Subtraction:Roentgenographic Differentiation Methods,published in 19348.

TomographyPlanigraphy, later to be known as tomography,was a form of body section imaging in which theimaging assembly was moved around a fixedpoint in the body. Structures lying in the sameplane as the rotation point would be imagedrelatively clearly, while structures lying above orbelow the plane would be blurred. Ziedses desPlantes claimed to have got the idea fromlooking into a glass of beer. The techniqueprovided clearer images of the skull base thanany previous method (Fig. 5).

Ziedses des Plantes’ Planigraph was taken

into commercial production by Massiot ofFrance, now part of Philips Medical Systems,and eventually became the Philips Polytome,which set the standard in tomography until theadvent of computed tomography (CT) in the1970’s.

Fig. 1. Plain skullradiograph showing acalcified tumour in thesellar region (arrow).

Fig. 2. Contrasttechniques.a. Pneumoencephalo-graphy: midlinetomography showing thefourth ventricle.b. Iodoventriculography.

1

2a

2b

158

Fig. 3. Myelography.a. Introduction of contrastmedium.b,c,d. Myelograms. Thepatient is anachondroplastic dwarf.

Fig. 4. Sagittaltomography in airmyelography.

Fig. 5. Large, partiallycalcified chondromyxoidtumour of the skull base.a. Plain radiograph.b,c. Two tomographicexposures from a series oflayers, showing thecalcified tumour in moredetail.

3a

3b

159

3c

3d

4

5a

5b

5c

160

Tomography was used to visualize thestructures of the skull and brain layer-by-layer.Linear tomography gave acceptable images(Fig. 6), but a pluridirectional movement suchas that of the Polytome gave better results (Fig. 7).

SubtractionSubtraction was based on the principle ofsubtracting one radiograph from another so thatonly the differences between them remained. Itsprincipal application was in angiography but,

although the principle had been described indetail in Ziedses des Plantes’ thesis in 1934, itremained little known until Ziedses des Plantespublished a monograph on the subject in 1960.By that time, the importance of angiography had become more widely recognized, and subtraction angiography soon became anestablished technique.

Originally, subtraction was performedphotographically (Fig. 8a) but by the end of the1960’s various television-based systems hadbecome available. In our hospital, we used sucha system (Evaluscope, Siemens) throughout the1970’s.

With the advent of digital techniques in theearly 1980’s, subtraction became generallyavailable in the form of Digital SubtractionAngiography (DSA) (Fig. 8b,c).

SomersaultingIn the 1960’s, the standard imaging equipmentfor neuroradiology was a tomograph with anisocentric chair.

The isocentric chair allowed the patient to berotated in a somersaulting technique, so that thewhole of the cerebral ventricles could bevisualized step-by-step using a relatively smallamount of air or gas. This fractionated techniquewas less stressful for the patient than using alarger amount of air to displace all of the fluidin the ventricles, but it remained uncomfortableand dangerous. In our hospital we used theMimer chair, which could convert from a bed,and served as a patient support for tomography.

The examination was routinely preceded byassessing the intracranial pressure via the opticfundus examination. Hypertension was alwaysa contraindication, as excessive pressure couldcause explosive decompression via the lumbarpuncture, causing the amygdala to constrict themedulla oblongata, resulting in the death of thepatient. If hypertension was detected, we wouldperform ventriculography with an oil- or water-soluble contrast medium, introduced directly viaa burrhole in the skull (Fig. 2b).

Pneumoencephalography was performed ona regular basis. We were fortunate in having theexcellent atlas published by Dr G. Ruggiero,an internationally acknowledged expert9. Theexamination was always performed undergeneral anaesthesia, using NO + O2 rather thanair, as this reduced the subsequent pain andnausea. Multiple tomography was performedwith the Mimer chair and a cassette with a stackof four or five films (Fig. 2a).

We used both gas cisternography and oilcisternography. If a tumour of the hypophysiswas suspected, gas was injected by the lumbarroute into the perisellar cisternae. The gas then

Fig. 6. Lineartomography.

Fig. 7. Pluridirectionaltomography.

6

7

161

surrounded the suprasellar cisternae and out-lined the tumour.

In the case of a suspected acusticusneurinoma, we would try to introduce gas intothe internal auditory canal. If the acusticusneurinoma was relatively small, so that it did not impede the passage of the gas, this approachwas generally successful. If not, we would followit up with an oil injection.

Nuclear medicineThe isotope scanner also came into widespreaduse in the mid 1960’s. As it was based on theuptake of labelled isotopes, it gave someindication of the brain metabolism, whiletumours might also show up as regions ofincreased activity.

In 1964, Di Chiro10 noted that intrathecalisotope injection could show the CSF flowdynamics. This was an important step forward,as anomalies in the flow dynamics can causebrain dysfunction.

Initially, images of isotope uptake wereobtained by rectilinear scanners, in which thepatient was scanned in parallel lines by amotorized detector. However, in 1963 Kuhl and

Edwards11 published details of a techniqueusing a computer, coupled with a transversesection imaging device, to produce transaxialtomograms of the head using single gamma rayradionuclides. This technique later becameknown as single photon emission tomography(SPECT).

By 1971 Kuhl and Sanders12 were usingtransverse section isotope scanning for thedifferential diagnosis of brain lesions.

Transverse sectional imaging yielded a wealthof information, but its value was limited by poorspatial resolution and its dependence on isotopeuptake (Fig. 9).

Fig. 8. Subtractionangiography.a. Photographicsubtraction.b. DSA image beforesubtraction.c. DSA image aftersubtraction.

Fig. 9. Radioisotopeimaging.

8a

8b 8c9

162

UltrasoundIn the early 1970’s, there was a growing interestin ultrasound. Its non-invasive nature made itparticularly attractive for paediatric appli-cations, especially in follow-up studies whererepeated exposure to X-rays would not bejustified.

Ultrasound was of value in detecting fetalabnormalities and for detecting anoxic lesionsin neonates and young children (Fig. 10). A-and B-mode echoencephalography could behelpful in the diagnosis of hydrocephalus, andpulsatile echoencephalography yielded charac-teristic results in various diseases.

The whole neonatal ventricular system couldbe visualized using the fontanelle as an acoustic

window.A-scan echoencephalography was also used

in the mid 1970’s for determination of midlineshift in both children and adults. Although thistechnique avoided X-ray exposure, it was notvery accurate, and was rapidly superseded bythe advent of computed tomography (CT).

In the late 1980’s, Levine used a duplexDoppler technique for cerebrovascular mappingin babies. Transosseous ultrasound could beused for the detection of ischaemia, and formonitoring subarachnoid haemorrhage.

Computed tomography (CT)In 1972, the whole discipline of neuroradiologywas revolutionized by the introduction of the

Fig. 10. Echography ofthe brain. Imagesreproduced by courtesy ofProfessor E.F. Avni.a. 7 day old baby.Coronal view showingfourth ventricle, temporallobes and posterior fossa.The hyperechoic whiteareas are blood,indicating a subarachnoidtemporal haemorrhage. b. Fetal head (samepatient as a). Parasagittalview showing fourthventricle and cerebellum.c. Head of 15 day oldbaby. Coronal viewshowing enlarged fourthventricle.d. Head of 15 day oldbaby (same patient as c).Parasagittal view showingenlarged fourth ventricleand cerebellum.

10a 10b

10c

IV

10d

163

CT scanner, developed by Godfrey N.Hounsfield at EMI Ltd. Hounsfield noted thatconventional X-ray techniques could notdistinguish between different tissues, as theyonly recorded the mean absorption of all thetissues through which the X-ray had penetrated.He reasoned that a system which performed setsof X-ray measurements through the body at amultitude of different angles would make moreefficient use of the information that the X-rayscould give, and might make it possible todistinguish between the various tissues of thebody13.

The original laboratory model used a gamma-ray source, and had an acquisition time of ninedays, with a reconstruction time of 21/2 hours.Nevertheless, the system worked, showing thatthe principle was sound.

A clinical prototype was built by EMI, andinstalled in Atkinson Morley’s Hospital,London, in 1972. The first patient scanned hada suspected brain lesion, and the image clearlyshowed a circular cyst in the brain,demonstrating that the machine was sensitiveenough to distinguish between normal anddiseased tissue.

In 1979, Hounsfield was awarded the NobelPrize in Medicine for his work on CT, togetherwith Allan Mc Cormack who had independentlyinvestigated the mathematical basis of CTreconstruction. In addition to many awards,Hounsfield received a knighthood in 1981.

The EMI scanner made it possible, for thefirst time, to obtain direct images of the softtissues of the brain. Primary tumours andmetastases, their origins, infiltrations and theirblood supply could all be seen in detail. At astroke, plain skull radiography, ventriculo-graphy, pneumoencephalography and much ofnuclear medicine had been superseded.

We decided to buy a whole-body CT scanner,and in 1976 a Delta Scanner (Ohio Nuclear)was installed in our department. An unusualfeature of this system was its ability to provideimages in colour. The image reconstructionsoftware had originally been devised forapplications in astronomy and satellite research,and had provision for false colour (Fig. 11).Although the images were undoubtedly ofartistic merit, the arbitrary allocation ofborderlines in graduated areas made littlecontribution to the diagnostic effectiveness.History repeated itself some years later in theintroduction of MR imaging (Fig. 12).

The Delta Scanner was painfully slow bypresent-day standards, taking 21/2 minutes forsimultaneous acquisition of two slices, and theslice thickness was 1 cm. Nevertheless, it was analmost incredible improvement on previous

11a

11b

12

Fig. 11. Early CT imagewith false colour.a. Axial head scan.b. Enlarged detail.

Fig. 12. Early MR imagewith false colour.

164

techniques. We could not only see the dis-placement of the ventricles very clearly, but wecould also see the exact location and extent ofthe tumours themselves (Fig. 13). Carefulselection of the grey scale and window levelsallowed us to take the first tentative steps in softtissue discrimination. Although the native CTimages yielded far more information than earliermethods, it was soon found that even betterresults could be obtained with the use of contrastmedium (Fig. 14).

The first CT scanners had no provision forscout views, so that it was not always easy torecognize the exact level in the body. Ourapproach to this problem was the use of a set ofradio-opaque rods of different lengths. Thenumber of rods in the image gave a guide to theslice position: the fewer the number of rods, themore distal the slice (Fig. 15).

By 1981, we had a CT scanner (Somatom,Siemens) providing thin slices that could bestacked for three-dimensional reconstruction(Fig. 16). Since then, computerized imageprocessing has given virtually unlimited scopefor CT angiography, reformatting in any

selected plane, and three-dimensional imagesgiving a surgeon’s eye view of pathologies.

Digital subtraction angiographyThe basic principles of subtraction had beendescribed by Ziedses des Plantes in 19348, but

Fig. 13. Early CT images.Although the quality ispoor by present-daystandards, the imageclearly shows theventricles, and also allowsdiscrimination of normaland pathological softtissues, representing aconsiderableimprovement over gas orcontrast cisternography.

Fig. 14. The effect ofcontrast medium in CT.a. Plain image.b. Image followinginjection of contrastmedium.

13 14a

14b

165

the technique was not generally applied until the1960’s, when interest was reawakened by thegrowing importance of angiography. Theoriginal photographic subtraction techniqueyielded relatively good results (Fig. 8a), but wastime-consuming and cumbersome.

By the end of the 1960’s, various television-based systems for subtraction had becomeavailable, and we used such a system(Evaluscope, Siemens) throughout the 1970’s.However, this system depended on theavailability of images on film.

Although the use of a balloon for occlusionof the cerebral vessels and carotid arteries

had been pioneered by Serbinenko in the late1960’s, and was described in the literature in197114,15, the absence of a satisfactory methodfor real-time angiography prevented wide-spread application of transluminal inter-ventional procedures.

With the advent of digital techniques in theearly 1980’s, subtraction became generallyavailable in the form of Digital SubtractionAngiography (DSA) (Fig. 8b,c). Thistechnique, originally referred to as ‘com-puterized fluoroscopy’, was developed by

Fig. 15. Early CT systemshad no provision for scoutviews. Radio-opaque rodsof different lengths give aguide to the slice position.The fewer the number ofrods, the more distal theslice.

Fig. 16. By 1981, thinslices could be stacked forthree-dimensionalreconstruction.a. Reconstruction of theskull of a child, less thanone year old, showingbone deficit associatedwith a large meningocelein the skull base.b. Compression fractureof C7 and luxation of C6-C7 with constriction ofthe vertebral canal.c. Reconstruction of avertebra following ananterior cervical graft,showing an unobstructedvertebral canal.

15

16a

16b

16c

166

Charles Mistretta and Andrew Crummy at theUniversity of Wisconsin16.

DSA provided real-time digital processingwithout intermediate analogue storage, makingit possible to view fully processed and enhancedsubtraction angiograms while the contrastmaterial is flowing through the region of interest.As a result, real-time angiograms both with andwithout subtraction are now available for rapidacquisition of diagnostic information, and for an ever-increasing number of interventionalapplications including balloon angioplasty,thrombolysis, and embolization with coils17.

Magnetic resonance imaging (MRI)At about the same time that we purchased ourfirst CT scanner, the first steps were taken in thedevelopment of MR imaging systems. Initially,the technique was referred to as NuclearMagnetic Resonance, as it depended on theproperties of the atomic nucleus, but the term‘nuclear’ was soon dropped, as patients were

sometimes worried by the incorrect butunderstandable association with radioactivity.

We were fortunate enough to be involved inthe development of the first commercialsystems, as we had been invited to co-operatewith Philips Medical Systems in thedevelopment of the Gyroscan range. In this way,our clinical requirements could be incorporatedright from the start.

It soon became clear that magnetic resonanceoffered some very specific advantages. It wasnon-invasive, and had an unprecedentedsensitivity to differences in tissue: the white andgrey matter were more clearly distinguished, and

Fig. 17. Prototype MRinstallation at the PhilipsMedical Systems factoryin Best, the Netherlands(author’s photograph).

Fig. 18. Modern MRimages.

17

18

167

some pulse sequences revealed pathologies thatcould not be seen in CT or with the standardMR pulse sequences. Moreover, there was acompletely free choice of image plane, includingexcellent midsagittal slices, with no need forreformatting.

The results obtained even with the firstexperimental system were so good that we weresoon bringing doubtful cases to the PhilipsMedical Systems factory in Best, theNetherlands, which is about 11/2 hours’ drivefrom our hospital in Brussels (Fig. 17). Thesewere mainly patients with a suspected braintumour which did not show up on CT.

The procedure was somewhat irregular, andgenerally involved the author driving patients tothe factory in her own car, as there was noadministrative provision for referring andtransporting patients to a factory. However, theresults obtained more than justified the logisticproblems, and the patients were more thanhappy to co-operate. Informed consent was notjust a formality.

Modern MR techniques (Fig. 18) are about100 times more sensitive than CT with respectto tissue discrimination, and are very fast, withthe possibility of performing a completeexamination in about four minutes. Protondensity weighting provides excellentvisualization of MS plaques, which can barelybe detected on the CT images. Appropriatechoice of pulse sequences can also show manybrain metastases which are not visible on CT.

Nowadays, no complex neurosurgicalprocedures are performed in our hospitalwithout prior MR examination. Patients arealways screened for brain metastases beforeresection of primary tumours, e.g. in the lung.

The specificity and sensitivity of MRexaminations can be enhanced still further withthe use of gadolinium contrast medium. This cangive valuable additional information on thepossible histology of a lesion, and some lesionsonly appear after contrast enhancement.

Examinations of the hypophysis showtumours better than repeated CT examinations,particularly as the hypophysis is too close to theskull base for satisfactory CT imaging. More-over, MR involves no radiation dose, which isparticularly important in this type ofexamination in view of the possibility ofexcessive exposure of the eyes.

The high sensitivity and lack of radiation ofMR examinations is of great value in paediatricapplications. The possibility of following brain maturation is of particular interest.Myelinization starts just before birth, andcontinues very actively through the first twoyears of life. A sequence of different changes

can be seen from month to month.MR angiography (Fig. 19) is relatively easy,

given a little experience and practice. Phasecontrast angiograms can be performed in about20 seconds, with no need for contrast injections,puncture or catheterization. As a consequence,the importance of conventional angiography has declined somewhat, as it is now restricted tothe diagnosis of strictly vascular lesions, ratherthan serving as an indirect sign of the presenceof a tumour.

MR spectroscopy has been performed in ourhospital on an experimental basis since 1985,under the direction of Professor Segebarth. Themain objective is to detect spectral changes inbrain tumours due to changes in the proportionsof the various metabolites and, if possible, toevaluate the possibility of using them as a basisfor differential diagnosis. Some clinics demandspectroscopy before surgery, but we find it lessspecific than we would like. However, it iscertainly useful for the detection of very earlyanoxic lesions, and for early detection ofmetabolic changes.

Abnormal metabolite levels could be apossible early sign of Alzheimer’s disease. Therehas been ongoing research into finding a specificMR sign for Alzheimer’s disease since 1985,notably that carried out by Dr Brian Ross andhis colleagues18.

The futureSpeculation about the future is alwaysdangerous. Sooner or later one’s forecasts willalways be compared with the reality, and all toooften there is something totally unexpectedlurking just around the corner. It wouldtherefore be absurd to even try to speculate onthe whole of the next century of diagnosticimaging. Nevertheless, extrapolation fromcurrent trends generally gives some idea ofdevelopments in the short term.

It is virtually certain that the trend towardscombined diagnosis and treatment willcontinue, with increasingly sophisticatedinterventional procedures for the treatment ofaneurysms, arteriovenous malformationsand/or stenoses.

Interventional techniques have been in usefor a quarter of a century14,15, but havedeveloped relatively slowly. For example, coilembolization is still being evaluated. On theother hand, endovascular treatment ofarteriovenous malformations is now a well-established alternative to surgery or, in somecases, the only available technique. We canconfidently predict increasing application ofballoon angioplasty, stenting and relatedtechniques in the near future.

168

Fig. 19. MR angiography.a. Circle of Willis.b. AVM. High-velocityflow-encoded imageshowing feeding arteries(arrows).c. AVM. Low-velocityflow-encoded imageshowing venous drainage(arrows).

d. AVM. Low-velocityflow-encoded image.Axial view showingvenous drainage.

19a 19b

19c

19d

169

There will also be more combinations ofimaging modalities (Fig. 20) as well ascombinations of imaging and measuringtechniques. Combinations of CT or MR withelectroencephalography, magnetoencephalo-graphy and nuclear medicine will yield newinsights into the functions of the living brain,while the combination of MR imaging and MRspectroscopy is already providing additionalinformation on the normal and pathologicalmetabolism.

Proton emission tomography (PET) showsmetabolic processes in vivo, indicating the sitesof brain activity such as response to visual

stimuli, as well as memory, which seems to takeplace in the same sites as the original response.However, PET is inherently slow, whereas MRIcan, in principle, provide results in milliseconds.

MR imaging will become increasinglyrefined. New pulse sequences such as FSE andTSE are continually being developed for fasterand better slice acquisition, includingsimultaneous multiple slice acquisition, whilewe are now only beginning to appreciate thediagnostic results obtainable with differentcombinations of pulse sequences. For example,it has long been known that T1 weighting cangive better differentiation of white and grey

Fig. 20. Combineddisplay. This examplecombines the bone imagefrom CT and the tumourimage from MR.

20

170

matter, while T2 weighting tends to be moresensitive to pathologies. It has also been foundthat the plaque associated with multiple sclerosisshows up better with 1H density weighting.Some tumours mimic normal tissue in the morecommonly used pulse sequences, but can berevealed by special pulse sequences such asFLAIR. The relationships between pulsesequences and pathologies will undoubtedly beexplored further in the coming years.

The combination of MR imaging andspectroscopy and, in particular, new MRprocedures for functional brain imaging such asbrain activation studies, perfusion imaging and

diffusion imaging, coupled with increasedcomputer power, should allow us to watch thebrain at work: literally as fast as thought.

Current computer techniques provide ever-faster reconstruction, as well as a vastassortment of image processing techniques,including reformatting in curved planes such asthe surface of an organ or vessel trajectories.Fast reconstruction in multiple layers offers newdiagnostic possibilities, such as cine excursionsthrough the vessels.

It is also possible to combine images fromseveral modalities in a single display (Fig. 20).

Stereotactic procedures with accuratelocalization using CT, MR and PET images arealready being used for precision biopsyprocedures and tissue implantation and arealso used for micro-interventions (Fig. 21).

Although neuroradiology has been aroundfor a century, the most impressive developments

have taken place within the last two decades,and the pace of development is acceleratingrapidly. We can therefore confidently predictthat, with the probable exception of W.C.Roentgen’s original stunning discovery,developments over the coming century will beeven more impressive than those of the lasthundred years.

References1. Nelson and Colne Express, 2 May 1896. Quoted inBull JWD and Fischgold H. A Short History ofNeuroradiology in: Cabanis EA, Cabanis M-T, eds.Contribution to the History of European Radiology, 2ndedition. Editions Pradel, Paris 1994: 15-16.2. Gutiérrez C. The Birth and Growth of Neuroradiologyin the USA. Neuroradiology 1981; 21: 227-237.3. Houser O.W. Neuroradiology: A HistoricalPerspective. Radiology 1995; 196: 1-2.4. Dandy WE. Ventriculography following the Injectionof Air into the Cerebral Ventricles. Ann Surg 1918; 68:5-11.5. Sicard JA, Forestier J. Méthode Generaled’Exploration Radiologique par l’Huile Iodée. Bull MemSoc Med Hop Paris 1922; 46: 463-469.6. Moniz E. L’Encephalographie Artérielle, sonImportance dans la Localisation des Tumeurs Cérébrales.Rev Neurol 1927; 2: 72-90.7. Moniz E. Cerebral Angiography. Paris. Masson 1931.8. Ziedses des Plantes BG. Planigraphie en Subtractie.Röntgenografische Differentiatie Methoden. Thesis,Utrecht 1934.9. Ruggiero G. L’Encephalographie Fractionée. Massoned. Paris 1957.10. Di Chiro G. Movement of the Cerebrospinal Fluid inHuman Beings (letter). Nature 1964; 204: 290-291.11. Kuhl DE, Edwards RQ. Image SeparationRadioisotope Scanning. Radiology 1963; 80: 653-662.12. Kuhl DE, Sanders TP. Characterizing Brain Lesionswith use of Transverse Section Scanning. Radiology 1971;98: 317-328.13. Hounsfield GN. Computed Medical Imaging. NobelLecture, December 1979. Journal of Computer AssistedTomography 1980; 4,5: 668-674.14. Serbinenko FA. Catheterization and Occlusion ofMajor Cerebral Vessels and Prospects for theDevelopment of Vascular Neurosurgery. Vopr Neirokhir1971; 35,5: 17-27. 15. Serbinenko FA. Occlusion of the Cavernous Portionof the Carotid Artery with a Balloon as a Method ofTreating Carotid-Cavernous Anastomosis. VoprNeirokhir 1971; 35, 6: 3-9.16. Mistretta CA, Ergun DL. The Development of DigitalSubtraction Angiography: Reflections from theUniversity of Wisconsin. MedicaMundi 1994; 39,1: 5-11.17. Pierot L, Boulin A, Castaings L, Moret J,Aerts JCJ. Current Techniques for EndovascularTreatment of Intracranial Aneurysms. MedicaMundi1994: 39,2: 80-93.18. Ross B. Abnormal Cerebral Metabolite Concen-trations in Patients with Probable Alzheimer’s Disease.Magn Res Med 1994; 32, 1: 110-115.

Fig. 21. Combination ofan optical 3D localizingsystem and a digitalworkstation allows foradvanced applications in(neuro)surgery. Theexample shown is thePhilips EasyGuideNeuro.

21