Journal Logo

Empirical Investigations

Preoperative Simulation for the Planning of Microsurgical Clipping of Intracranial Aneurysms

Marinho, Paulo MD; Vermandel, Maximilien PhD; Bourgeois, Philippe MD; Lejeune, Jean-Paul MD, PhD; Mordon, Serge PhD; Thines, Laurent MD, PhD

Author Information
Simulation in Healthcare: Journal of the Society for Simulation in Healthcare: December 2014 - Volume 9 - Issue 6 - p 370-376
doi: 10.1097/SIH.0000000000000056

Modern imaging techniques such as multiple-detector 3-dimensional computed tomography angiography (3D-CTA), 3D digital subtraction angiography (3D-DSA), and 3D magnetic resonance imaging (3D-MRI) permit very precise depiction of the intracranial vasculature and of the surrounding environment, which includes 2 main components, namely, bone (the skull base and cranial vault) and the brain parenchyma.1–21

Various factors affect the success of surgery for intracranial aneurysms (IAs). One of these factors is the ability of the neurosurgeon to visualize an accurate 3D representation of the surgical scene to anticipate potential risks or difficulties in exposing and clipping the aneurysm (ie, by using the proper surgical approach to locate the rupture site). This is particularly important in the management of ruptured IAs because of the high risk of intraoperative bleeding. The surgeon generally relies on the mental combination of 3D reconstructions from multiple imaging modalities to accurately analyze the vascular anatomy or aneurysm morphology and choose the safest approach to the aneurysm. Consequently, the quality of the microsurgical treatment of an IA depends on the association of strict preoperative planning (for which surgical simulation is potentially useful), a continuously updated intraoperative strategy (for which neuronavigation tools are potentially useful), and expertise in cerebrovascular surgery (for which specific training in cerebrovascular surgery is useful).

In a recent article reviewing the available tools for the simulation of IA clipping or cerebrovascular neuronavigation,22 we found that most of the simulated procedures did not thoroughly display the essential visual aspects of the surgical approach, particularly the environment surrounding the cerebral arteries (ie, the cranial bone and brain parenchyma), or were not designed to simulate the surgical approach together with the positioning of the clip on the aneurysm neck. Therefore, the purpose of this study was to evaluate the technical feasibility and the clinical relevance of a new method of simulation of IA clipping using a simulation protocol combining a complete 3D model of the surgical scene with a visual representation of the clip application.

METHODS

Segmentation of the Cranial Bone, Brain Cortex, Cerebral Arteries, and Aneurysm Clips

The first modeling step was performed using the free Osirix DICOM image processing application (Pixmeo, Geneva, Switzerland), which allowed the segmentation and merging of the 3 main structures of interest: the cranial bone, brain cortex, and cerebral arteries (Video 1, Supplemental Digital Content 1, http://links.lww.com/SIH/A157, which demonstrates the steps of segmentation). Native images of the cranial bone and cerebral arteries were acquired from a preoperative 3D-CTA scan with iodine contrast injection performed on a Siemens Healthcare Sensation 64-detector scanner (Siemens Medical Solutions, Malvern, PA). The 3D reconstruction of the bone structures was performed using a simple thresholding segmentation method (Fig. 1A). The fashioning of the desired pterional (frontotemporal) craniotomy was performed manually using the cutting tool in the software and was designed to correctly expose the aneurysm: frontal extension for anterior communicating artery aneurysms and temporal extension for middle cerebral artery (MCA) or internal carotid artery aneurysms. The 3D reconstruction of the intracranial arteries (Fig. 1C) was performed using the CMIV CTA plug-in23 for Osirix, which was originally developed for the processing of CTA images of the coronary arteries and uses an algorithm based on a competing fuzzy connectedness tree to perform a skeletonization of the arteries.24 Native images of the brain parenchyma were acquired using a preoperative T1-weighted 3D-MRI scan of the brain with contrast enhancement on a 1.5-T MRI scanner (Achieva, Philips Medical Systems, Best, the Netherlands). The 3D reconstruction of the brain cortical surface (Fig. 1B) was performed separately and semiautomatically using the region-growing segmentation algorithm in Osirix and operator-defined points (ie, seeds) within the regions of interest. We also had to take into account that during the clipping of an IA, some cerebral retraction is needed to correctly expose the vascular malformation. To visually represent this brain deformation induced by the operator during surgery, we performed an artificial corticectomy around the aneurysm in the 3D model of the brain (Fig. 1B). Finally, a commonly used set of miniature and standard temporary aneurysm clips (B. Braun-Aesculap, Tuttlingen, Germany), fixed in their forceps, was digitized using the same 3D-CT scanner and then segmented with Osirix using a simple thresholding method (Fig. 1D).

FIGURE 1
FIGURE 1:
Native images from MRI and CTA Scans were imported into Osirix and then segmented separately to provide 3D reconstructions of the skull (A), the brain (B) and the intracranial arteries on the operating side (C). A virtual craniotomy (A) was performed on the cranial model based on the aneurysm location and the desired surgical approach. A virtual corticectomy (B) was performed on the cerebral model at the level of the predicted location of the brain retraction that would be necessary during surgery to expose the aneurysm. These 3D reconstructions were then merged to build the final model of the surgical scene (comprising the skull, brain, and arteries), which was imported into Blender together with the digitized clips and forceps (D) to enable the visual simulation of the aneurysm clipping procedure.

Modeling of the Surgical Scene

The 3D models of the cranial bone and brain were then merged using the cerebral arterial tree as a reference image (ie, the image on which the other sets of images were superimposed) to build the complete surgical scene representing the vascular exposure (Video 1, Supplemental Digital Content 1, http://links.lww.com/SIH/A157, which demonstrates the steps of surgical scene modeling). This step required the use of the iterative closest point (ICP) algorithm in Osirix after the reference points in each exam were selected and entered manually using the point-based registration function before the matching process. To avoid excessive superimposition of bone or arterial structures from the opposite side and to improve the view through the craniotomy, only hemi-bone and hemi-vascular 3D models were used.

Visual Simulation of Aneurysm Clipping with Blender

For this step of the modeling, we used free Blender software (NV2 Sarl, Paris, France) to enable the simulation of the insertion and mobilization of the aneurysm clips and their appliers in the surgical scene (Video 1, Supplemental Digital Content 1, http://links.lww.com/SIH/A157, which demonstrates how the clip forceps were manipulated in the surgical scene to visually simulate the aneurysm clipping process). Blender is a modeling platform frequently used to create 3D animations, which enables the importation of multiple segmented structures, which can then be visually manipulated. The previously constructed models of the surgical scene and the clips fixed in their appliers were exported from Osirix into the Blender in the Virtual Reality Markup Language (VRML) format using the 3D surface rendering function. During the simulation of the surgical procedure, the surgical scene could then be rotated in a 3D space by the neurosurgeon to reproduce the view that he or she would expect to see during the procedure. At this stage, the surgeon could introduce different clips, fixed in their appliers, into the operative field. The process of clipping the aneurysm by orienting and positioning the clip along the aneurysm neck was visually simulated (Fig. 2 and Fig. 2bis color, Supplemental Digital Content 2, http://links.lww.com/SIH/A158). This allowed the surgeon to plan the number, size, shape, and orientation of clips that would be appropriate according to the vascular anatomy and to anticipate the possible interaction between the clip appliers and the surrounding bony or cortical surfaces. The neurosurgeon also had the opportunity to prepare for the first steps of the intervention by deciding where to start the opening of the arachnoid cisterns, planning different sequences of arterial tree exposition, and detecting the potential risks of the aneurysm exposure according to the aneurysm’s orientation, vascular relationships, and site of rupture (if indicated).

FIGURE 2
FIGURE 2:
Comparison of the visual simulation and the operative procedure in a patient treated for an aneurysm of the right MCA. Excellent similarity between the visual representation of the surgical scene obtained using Blender (A) and the aspect of the operative field before complete aneurysm exposure (B). A*, Global view of the head orientation in the visual simulation. C and D, Excellent similarity in the appearance of the aneurysm between the visual simulation and the operative view. The clip size, shape, and orientation selected during the simulation were identical to the clip chosen during surgery (C). Cotonoids have been applied on the frontal and temporal cortical surfaces to avoid any injury to the brain while manipulating the instruments in the operative field. E and F, Excellent similarity in the clip shape, size, and final orientation between the visual simulation of the clipping process and the actual clip application (F*). To achieve appropriate clip positioning, the aneurysm sac was temporarily tilted to the opposite side in the largest intraoperative image.

Clinical Validation

To assess the feasibility and estimate the potential usefulness of the simulation in clinical conditions, the simulation protocol was followed before the clipping of 10 IAs in the anterior circulation (9 unruptured) in 6 patients referred for microsurgical treatment. After their informed consent was obtained, all patients underwent preoperative cerebral 3D-MRI and 3D-CTA, from which a patient-specific 3D surgical scene was generated based on the recommendations of the involved neurosurgeon. Screenshots of the visual simulation of the clipping process were saved for comparison with intraoperative views. Two neurosurgeons evaluated the results of the simulation protocol; the first surgeon (L.T.) had been board certified in neurosurgery for 8 years, and the second surgeon (P.B.) had been board certified in neurosurgery for 20 years. They analyzed 7 aneurysms (2/3) and 3 aneurysms (1/3), respectively.

For each case, the following parameters were preoperatively recorded:

  • The aneurysm characteristics (location, side, and neck size)
  • The applicability of the protocol
  • The visual quality (poor, average, good, or excellent) of the simulation in the representation of the surgical scene and the depiction of the vascular anatomy
  • The planning of the first steps of the surgical approach: location of the craniotomy, site of arachnoid cistern opening, and first vascular structure to be exposed
  • The selection of the optimal characteristics of the clip(s) to obtain complete aneurysm occlusion: number, shape, size, and orientation

For each case, the following parameters were intraoperatively recorded:

  • The accuracy of the simulation procedure (poor, average, good, or excellent): visual similarity between the surgical scene and the main operative view, accurate preoperative identification of the vascular structures (ie, the diseased artery and the collateral arteries) and their relationship to the aneurysm, and appropriate guidance of the surgical approach
  • Clipping procedure parameters: the degree of similarity between the characteristics (number, shape, size, and orientation) of the clip(s) applied during surgery and the clip(s) selected during the simulation, as well as the number of clipping attempts.

The feasibility of the procedure was evaluated according to the ability to apply the entire simulation protocol to each aneurysm case and according to the analysis of the surgical scene parameters, namely, the visual quality, the similarity with the operative views, and the anatomic accuracy compared with the actual operation in the description and interpretation of the vascular anatomy.

The utility of the simulation protocol was assessed according to its ability to appropriately support the neurosurgeon’s decision-making process during surgical planning, namely, the prediction of the first steps of the surgical approach and the choice of the most appropriate number, shape, size, and orientation of the clip(s). The utility of the protocol was also evaluated according to its influence in achieving the most straightforward operation by reducing aneurysm manipulation during clipping.

RESULTS

All the patients underwent microsurgical clipping of their aneurysm(s) without operative or neurologic complications, except in 1 patient who experienced reversible seizures postoperatively. Complete occlusion was obtained for all aneurysms (6 MCA, 2 anterior communicating artery, and 2 internal carotid artery aneurysms; average neck size, 4 mm) and confirmed intraoperatively by indocyanine green videoangiography25 and/or aneurysm dome puncture and postoperatively by 3D-CTA or 3D-DSA. The simulation protocol could be applied for all patients, resulting in appropriate surgical planning.

At the preoperative stage, the visual quality of the simulation was rated good or excellent for the representation of the surgical scene in 4 and 5 cases, respectively. In one case, the presence in the CTA data of artifacts from a previous endovascularly treated associated IA precluded the optimal segmentation of the surrounding structures. The preoperative analysis of the vascular anatomy (ie, the pixelated views of the vessels in Blender) was rated good in 10 cases. The 3D reconstructions were helpful in every case in planning the first steps of the surgical approach, namely, determining the size and location of the craniotomy, the site of the arachnoid cistern opening, and the first vascular structure to be exposed. On the basis of the preoperative simulation, the neurosurgeon was also able in all the aneurysms to anticipate the characteristics (ie, the number, size, shape, and orientation) of the clip(s) that would be required to insure their complete occlusion. One clip was deemed necessary in 9 aneurysms and 2 clips in a large bilobulated MCA aneurysm.

At the operative stage, the visual similarity between the model of the surgical scene and the main operative view used during the surgical procedure was rated excellent in all cases. The accuracy of the simulation for the preoperative identification of the important vascular structures surrounding the aneurysm was excellent in all cases but one (which was rated poor) as confirmed by the intraoperative findings. Because of its very small caliber, the origin of 1 anterior choroidal artery (vascularizing the motor tract through the internal capsule) was not visible at the neck of a small posterocarotid aneurysm in the preoperative CTA data and, as a consequence, in the 3D model. The neurosurgeons found the overall simulation helpful in all cases (a good rating) for the guidance of the surgical approach (ie, the fashioning of the craniotomy, the opening of the arachnoid cisterns, and arterial tree exposure). The number of clip(s) applied during surgery was similar to the number of clip(s) preoperatively predicted in simulation in 9 cases, including an aneurysm that required 2 clips. In one case, because of the unpredictable thickness of the aneurysm wall (due to atherosclerosis), the complete occlusion of the aneurysm lumen could only be obtained by the application of an identical supplementary clip. The characteristics (shape, size, and orientation) of the ideal clip(s) were preoperatively predicted during simulation and subsequently confirmed during surgery by the application of the same clip in 8 aneurysms. In one case, during the operation, it was decided to use a longer clip (of the same shape); in a second case, a completely different clip (in size, shape, and orientation) was used. In total, a straightforward clip application (with only one attempt needed) was performed in 6 aneurysms using the disposable clip that was preoperatively selected in the simulation. The final clips were then applied to replace the temporary clips.

The overall feasibility and utility of the simulation protocol are summarized in Table 1.

TABLE 1
TABLE 1:
Feasibility and Utility of the Visual Simulation Protocol for IA Clipping

DISCUSSION

Previous Studies on Simulation for IA Clipping

High-resolution 3D imaging techniques have considerably influenced the way IAs are treated currently by dramatically improving the understanding of the angioarchitecture of the aneurismal sac (ie, its shape, neck morphology, and site of weakness) and of the neighboring anatomic relationships (ie, the collateral arteries, brain parenchyma, and nerves). The development of 3D-CTA and 3D-DSA protocols dedicated to determining the orientation of the vascular tree as it would be seen during surgery has aided the planning of these interventions because the images facilitate the understanding of the 3D presentation of the aneurysm before and during surgery and allow the neurosurgeon to anticipate potential difficulties in the clipping procedure.16,18,26 Nevertheless, some authors understandably found it insufficient to have only the possibility of visualizing an isolated 3D arterial reconstruction through a craniotomy without being able to interact with the images or to simulate the clipping process itself.27 These authors evaluated the utility of the Dextroscope software (Volume Interactions, Singapore) for IA surgery. This is an expansive virtual-reality simulator for the planning and training of operative neurosurgical procedures. The authors loaded patients’ 3D-CTA and cranial bone 3D-CT scans as well as digitized clips and appliers, into the Dextroscope workstation. A craniotomy was first performed using the drill function of the software. The vascular tree and the aneurysm were then oriented according to the expected operative view. Finally, the user was able to select the appropriate clip and to practice clipping by sliding the virtual clip along the aneurysm neck. This simulation tool allowed the user to fashion the ideal craniotomy to obtain the optimal and most realistic surgical exposure of the aneurysm. It was also helpful in choosing the most suitable clip shape, size, and orientation. Other authors have focused their efforts on developing virtual or hollow (solid or elastic) models of the cerebral vasculature of patients with IAs.28–31 These models were used to estimate the deformability of the aneurysm under clip application, to improve the 3D understanding of the aneurysm anatomy or to select and practice the optimal clip approach based on the morphology of the aneurysm and collateral vessels.

Although these studies showed that it was possible to simulate either the surgical exposure or the final clip application using high-resolution multimodality 3D imaging, they did not provide a protocol integrating all the necessary steps. In addition, although the 3D rendering of the vascular tree is appealing, these techniques lack haptic operator controls to transmit to the user a sense of the tissue strength and force feedback from the simulated clip applier. Another drawback of these methods is the absence of information about the thickness or the biomechanical properties of the aneurysm wall and the parent vessels, which would be useful to predict their deformability during clipping. Indeed, the effect of the clip application on the aneurysm sac might dramatically differ according to these parameters, resulting either in a perfect arterial wall repair or in arterial deformation leading to lumen occlusion and stroke. More importantly, the techniques previously described lack the depiction of the surrounding brain surface, which is one of the main structures restricting the accessibility (ie, the surgical corridor) to the aneurysm with the instruments (ie, the clip appliers).

Present Study on Visual Simulation of IA Clipping

This study aimed to evaluate the technical feasibility and the clinical utility of a simulation procedure for IA clipping that integrated the 3D modeling of the surgical scene with the visual representation of clip application. To perform these 2 tasks, we combined the use of 2 already available applications, Osirix and Blender, to develop a practical and useful solution. For the first time, parenchymal data were introduced into the procedure to represent the surgical corridor, with cerebral retraction around the aneurysm being approximated by the creation of a virtual corticectomy on the 3D model of the brain. More importantly, we have technically validated the feasibility of the method in clinical conditions in a series of 10 IAs by showing in each case the excellent similarity, before and during clipping, between the simulation of the surgical scene and the actual operative view. The clinical utility of the simulation was confirmed in the majority of the cases by the ability of the neurosurgeon to plan the intervention appropriately (ie, the fashioning of the craniotomy, the opening of the arachnoid cisterns, and arterial tree exposure), to detect the potential risks of the surgery and to select the appropriate clip. An immediate application of this tool would be the preoperative evaluation of the feasibility and safety of minimally invasive procedures that are currently becoming more widely used for IA clipping.26,32–34 These approaches require strict planning, focusing on the accessibility and presentation of the aneurysm through the preselected narrow operative corridor, with particular attention paid to potential interactions between the instruments and the cranial and cerebral structures.

If the concept of the simulation of IA clipping has been technically validated in this work, its clinical impact should be explored in a larger series of IA operations performed with and without the described procedure and by different operators. The assessment of this simulation protocol was only performed in our study by a small panel of neurosurgeons, which could introduce a high risk of analysis bias (ie, subjectivity). Obviously, this study was a preliminary evaluation of the clinical potential of such a tool and should be followed by a large set of trials to determine whether practice on the simulator really improves measures such as the duration and accuracy of the intervention, the complication rate, the number of clips used or wasted, and the number of attempts needed to place a clip. In addition, this simulation procedure remains relatively primitive because no integrated application has been designed yet to allow the automatic processing of the 3D imaging data and the simultaneous modeling of the surgical scene to simulate the aneurysm clipping procedure. All the required steps were performed manually or semiautomatically, and the manipulation of the clip appliers in Blender was not sufficiently intuitive. Imaging artifacts are also a possible limitation of this simulation technique, as we experienced in one case.

In many respects, this simulation procedure primarily represents a method for preoperative planning rather than an actual simulation of the surgery because of the impossibility of directly interacting with the surgical scene. The proposed model offers only a partial depiction of the final clipping process because it only represents the superimposition of the clip along the aneurysm neck without providing a rendering of the aneurysm wall behavior under the clip application or operator haptic feedback from the clip applier. Consequently, using this simulation does not provide any training advantage to the user, at least in the physical skills required. Nevertheless, this visual simulation procedure seems to be a step toward the creation of a true operative simulator. In the future, it will be necessary to integrate the 3D surgical scene used in this study with more sophisticated capabilities to enhance the physical and sensory experience of the operator for greater realism. The patient’s specific biomechanical characteristics will also have to be taken into account to accurately reproduce the deformation of the brain and aneurysm under moderate cerebral retraction and clip application, respectively. Ideally, haptic operator controls will be included to transmit to the user the sense of the tissue strength and force feedback from the simulated clip applier.31,35–37 This will make this protocol suitable not only for operative planning but also for surgical training. This approach will become still more relevant for aneurysm clipping simulation as more aneurysms are treated using endovascular techniques and surgical management becomes less frequent.

CONCLUSIONS

This study showed that the simulation of IA clipping based on the processing of high-resolution 3D imaging data could become effective. This is a new and important step toward the design of a more sophisticated simulation platform that would integrate a high-quality 3D model of the surgical scene, biomechanical properties of the brain and aneurysm, and precise operator haptic feedback. The development of a patient-specific cerebrovascular surgery simulator based on these requirements now seems more realistic and achievable.

REFERENCES

1. Beck J, Rohde S, Berkefeld J, et al. Size and location of ruptured and unruptured intracranial aneurysms measured by 3-dimensional rotational angiography. Surg Neurol 2006; 65: 18–25.
2. Baxter AB, Cohen WA, Maravilla KR. Imaging of intracranial aneurysms and subarachnoid hemorrhage. Neurosurg Clin N Am 1998; 9: 445–462.
3. Bertalanffy H, Gilsbach JM, Mayfrank L, et al. Planning and surgical strategies for early management of vertebral artery and vertebrobasilar junction aneurysms. Acta Neurochir (Wien) 1995; 134: 60–65.
4. Boor S, Resch KM, Perneczky A, et al. Virtual endoscopy (VE) of the basal cisterns: its value in planning the neurosurgical approach. Minim Invasive Neurosurg 1998; 41: 177–182.
5. Dietrich J, Gunther L, Fried H. [3-dimensional computerized tomography in diagnosis and surgical planning of intracranial aneurysms]. Zentralbl Neurochir 1995; 56: 34–39.
6. Feng H, Tan H, Kiya K, et al. Three-dimensional CT angiography and surgical correlation in the evaluation of intracranial aneurysms. Chin Med J (Engl) 2002; 115: 1146–1149.
7. Harbaugh RE, Schlusselberg DS, Jeffery R, et al. Three-dimensional computed tomographic angiography in the preoperative evaluation of cerebrovascular lesions. Neurosurgery 1995; 36: 320–326.
8. Heffez DS, Mikhael M, Jensen K. Operative confirmation of three-dimensional computed tomographic and magnetic resonance imaging of cerebrovascular pathology. J Image Guid Surg 1995; 1: 179–190.
9. Kallmes DF, Kallmes MH, Lanzino G, et al. Routine angiography after surgery for ruptured intracranial aneurysms: a cost versus benefit analysis. Neurosurgery 1997; 41: 629–639.
10. Karamessini MT, Kagadis GC, Petsas T, et al. CT angiography with three-dimensional techniques for the early diagnosis of intracranial aneurysms. Comparison with intra-arterial DSA and the surgical findings. Eur J Radiol 2004; 49: 212–223.
11. Kato Y, Sano H, Katada K, et al. Usefulness of simulation of surgical approaches to cerebral aneurysms by helical scanning CT (HES-CT). Minim Invasive Neurosurg 1995; 38: 99–104.
12. Narisawa A, Syamoto H, Karibe H, et al. [MRA volume rendering for surgical planning of unruptured intracranial cerebral aneurysms]. No Shinkei Geka 2005; 33: 243–248.
13. Nishihara M, Tamaki N. Usefulness of volume-rendered three-dimensional computed tomographic angiography for surgical planning in treating unruptured paraclinoid internal carotid artery aneurysms. Kobe J Med Sci 2001; 47: 221–230.
14. Park J, Hamm IS. Anterior interhemispheric approach for distal anterior cerebral artery aneurysm surgery: preoperative analysis of the venous anatomy can help to avoid venous infarction. Acta Neurochir (Wien) 2004; 146: 973–977.
15. Satoh T. Transluminal imaging with perspective volume rendering of computed tomographic angiography for the delineation of cerebral aneurysms. Neurol Med Chir (Tokyo) 2001; 41: 425–429.
16. Siablis D, Kagadis GC, Karamessini MT, et al. Intracranial aneurysms: reproduction of the surgical view using 3D-CT angiography. Eur J Radiol 2005; 55: 92–95.
17. Suzuki Y, Nakajima M, Ikeda H, et al. Preoperative evaluation of the venous system for potential interference in the clipping of cerebral aneurysm. Surg Neurol 2004; 61: 357–364.
18. Thines L, Taschner C, Lejeune JP, et al. Surgical views from three-dimensional digital subtraction angiography for the planning of aneurysm surgery. J Neuroradiol 2007; 34: 205–211.
19. Watanabe Z, Kikuchi Y, Izaki K, et al. The usefulness of 3D MR angiography in surgery for ruptured cerebral aneurysms. Surg Neurol 2001; 55: 359–364.
20. Taschner CA, Thines L, Lernout M, et al. Treatment decision in ruptured intracranial aneurysms: comparison between multi-detector row CT angiography and digital subtraction angiography. J Neuroradiol 2007; 34: 243–249.
21. Sugahara T, Korogi Y, Nakashima K, et al. Comparison of 2D and 3D digital subtraction angiography in evaluation of intracranial aneurysms. AJNR Am J Neuroradiol 2002; 23: 1545–1552.
22. Marinho P, Thines L, Verscheure L, et al. Recent advances in cerebrovascular simulation and neuronavigation for the optimization of intracranial aneurysm clipping. Comput Aided Surg 2012; 17: 47–55.
23. Wang C, Smedby O. Integrating automatic and interactive methods for coronary artery segmentation: let the PACS workstation think ahead. Int J Comput Assist Radiol Surg 5: 275–285.
24. Wang C, Smedby O. Coronary artery segmentation and skeletonization based on competing fuzzy connectedness tree. Med Image Comput Comput Assist Interv 2007; 10: 311–318.
25. Raabe A, Nakaji P, Beck J, et al. Prospective evaluation of surgical microscope-integrated intraoperative near-infrared indocyanine green videoangiography during aneurysm surgery. J Neurosurg 2005; 103: 982–989.
26. Mori K, Esaki T, Yamamoto T, et al. Individualized pterional keyhole clipping surgery based on a preoperative three-dimensional virtual osteotomy technique for unruptured middle cerebral artery aneurysm. Minim Invasive Neurosurg 54: 207–213.
27. Wong GK, Zhu CX, Ahuja AT, et al. Craniotomy and clipping of intracranial aneurysm in a stereoscopic virtual reality environment. Neurosurgery 2007; 61: 564–568 discussion 568–569.
28. D’Urso PS, Thompson RG, Atkinson RL, et al. Cerebrovascular biomodelling: a technical note. Surg Neurol 1999; 52: 490–500.
29. Kimura T, Morita A, Nishimura K, et al. Simulation of and training for cerebral aneurysm clipping with 3-dimensional models. Neurosurgery 2009; 65: 719–725 discussion 725–726.
30. Koyama T, Hongo K, Tanaka Y, et al. Simulation of the surgical manipulation involved in clipping a basilar artery aneurysm: concepts of virtual clipping. Technical note. J Neurosurg 2000; 93: 355–360.
31. Koyama T, Okudera H, Kobayashi S. Computer-assisted geometric design of cerebral aneurysms for surgical simulation. Neurosurgery 1995; 36: 541–546.
32. Mori K, Yamamoto T, Nakao Y, et al. Lateral supraorbital keyhole approach to clip unruptured anterior communicating artery aneurysms. Minim Invasive Neurosurg 2008; 51: 292–297.
33. Mori K, Osada H, Yamamoto T, et al. Pterional keyhole approach to middle cerebral artery aneurysms through an outer canthal skin incision. Minim Invasive Neurosurg 2007; 50: 195–201.
34. Kitano M, Taneda M. Extended transsphenoidal approach to anterior communicating artery aneurysm: aneurysm incidentally identified during macroadenoma resection: technical case report. Neurosurgery 2007; 61: E299–E300; discussion E300.
35. Koyama T, Okudera H, Kobayashi S. Computer-generated surgical simulation of morphological changes in microstructures: concepts of “virtual retractor.” Technical note. J Neurosurg 1999; 90: 780–785.
36. Wang P, Becker AA, Jones IA, et al. A virtual reality surgery simulation of cutting and retraction in neurosurgery with force-feedback. Comput Methods Programs Biomed 2006; 84: 11–18.
37. De Paolis LT, De Mauro A, Raczkowsky J, et al. Virtual model of the human brain for neurosurgical simulation. Stud Health Technol Inform 2009; 150: 811–815.
Keywords:

Simulation; Intracranial aneurysm; Image-guided surgery; Neurosurgical planning; Cerebrovascular surgery; Minimally invasive surgery

Supplemental Digital Content

© 2014 Society for Simulation in Healthcare